{"items":[{"id":"embeddings/pinecone","slug":"pinecone","category":"embeddings","access":"unconditional","accessBucket":"free","name":"Pinecone","provider":"Pinecone","planName":"freemium","kind":"Managed vector database","description":"Fully managed serverless vector database for similarity search and RAG, with built-in embedding and reranking inference.","logo":"/logos/pinecone-builder.png","url":"https://www.pinecone.io/","freeLimit":"The Starter plan is free with no credit card: up to 2 GB storage, 2M write units/mo, 1M read units/mo, and up to 5 serverless indexes on AWS us-east-1, plus 5M embedding tokens/mo of Inference and 1 GB of Assistant storage.","bestFor":"Solo developers prototyping semantic search, RAG chatbots, and recommendation engines without managing vector-search infrastructure.","caveats":"Starter is locked to AWS us-east-1 and a single project (max 2 users); usage beyond the included limits is blocked rather than billed, so production scaling requires the paid Builder ($20/mo) or Standard ($50/mo min) plans. Reranking is restricted to the bge-reranker-v2-m3 model only.","sourceUrl":"https://www.pinecone.io/pricing/","highlights":["Genuinely free with no credit card required — usage is capped/blocked, not charged","Up to 5 serverless indexes with 2 GB storage, 2M write units and 1M read units per month","Includes 5M embedding tokens/mo of built-in Inference (no separate embedding provider needed)","Bundled Assistant with 1 GB storage and monthly token allowances for RAG apps"],"tags":["vector-database","rag","semantic-search","embeddings","serverless","ai","similarity-search","managed"],"pricingModel":"freemium","renewal":"monthly","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"Database storage is capped at up to 2 GB on the Starter plan.","label":"2 GB storage cap","renewable":false,"sentiment":"neutral","severity":"medium","type":"single_resource_cap","unit":"GB","value":2},{"detail":"Up to 2M write units per month; resets monthly and additional usage is blocked, not billed.","label":"2M write units/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"M write units/mo","value":2},{"detail":"Up to 1M read units per month; resets monthly and additional usage is blocked, not billed.","label":"1M read units/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"M read units/mo","value":1},{"detail":"Up to 5 serverless indexes, each limited to 100 namespaces.","label":"Max 5 indexes","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"indexes","value":5},{"detail":"Starter indexes can only be created on AWS in the us-east-1 region; choosing cloud/region requires a paid plan.","label":"AWS us-east-1 only","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Starter is limited to a single project and up to 2 users.","label":"1 project, max 2 users","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"project","value":1},{"detail":"Built-in Inference includes 5M tokens/mo per embedding model (llama-text-embed-v2, multilingual-e5-large, pinecone-sparse-english-v0); resets monthly.","label":"5M embedding tokens/month (Inference)","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"M tokens/mo","value":5},{"detail":"Inference reranking on Starter is restricted to bge-reranker-v2-m3 only, with 500 requests/mo included.","label":"Reranking limited to one model","renewable":true,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"requests/mo","value":500},{"detail":"Assistant includes 1 GB storage, 500k input tokens/mo (promo: 1M/mo until June 30, 2026), 300k output tokens/mo, 500k context-processed tokens/mo, and 1,000 ingestion units/mo (USA region only).","label":"Assistant 1 GB + token allowances","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"GB","value":1},{"detail":"Starter plan support is community-based via Discord; paid plans add free/pro support with SLAs.","label":"Community support only","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"Backup/restore, import from object storage, dedicated read nodes, user/API-key RBAC, SAML SSO, and Prometheus/Datadog monitoring are not available on Starter.","label":"No backups, RBAC, SSO, or monitoring","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"}],"keyMetrics":[{"display":"5M tokens/mo","key":"free_tokens_per_month","label":"Monthly tokens","primary":true,"unit":"tokens/month","value":5000000},{"display":"1 tokens/min","key":"tokens_per_minute","label":"Tokens/min","note":"policy_default: Conservative scoring fallback: no comparable token throughput allowance is published in this item.","primary":false,"unit":"tokens/minute","value":1},{"display":"2 GB","key":"storage_gb","label":"Storage","primary":false,"unit":"GB","value":2}],"verified":true,"timestamp":"2026-06-17T00:00:00+00:00","badges":["🟢 Free forever","🟢 Commercial OK","🟠 Single resource","🟢 Embeddings/RAG"],"valueModel":{"bucket":"free","calc":{"allowance_usd_mo":0.7,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":5.1,"unit":"score","value":5.1,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.1,"b":3,"formula":"fixed_log_v1","key":"capacity","label":"Free value","provenance":"provider_quantity_times_reference_price","score":2.7,"unit":"$/mo","value":0.7},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Search fit","provenance":"derived_from_limitations_badges_and_access_terms","score":7,"unit":"score","value":7},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Access","provenance":"derived_from_limitations_badges_and_access_terms","score":8.3,"unit":"score","value":8.3}],"required_axis_count":3,"score":5.1,"score_formula":"weighted_components_v1","value":0.7,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"token_month","source":"frontmatter_keyMetrics","unit":"tokens/month","value":5000000},{"key":"token_minute","source":"frontmatter_keyMetrics","unit":"tokens/minute","value":1},{"key":"db_storage_gb_month","source":"frontmatter_keyMetrics","unit":"GB","value":2}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"embeddings/zilliz-cloud","slug":"zilliz-cloud","category":"embeddings","access":"unconditional","accessBucket":"free","name":"Zilliz Cloud","provider":"Zilliz Cloud","planName":"freemium","kind":"Managed Vector DB","description":"Fully managed, Milvus-powered vector database for AI, RAG, and semantic search, with a perpetual free tier.","logo":"/logos/zilliz-cloud-serverless.png","url":"https://zilliz.com","freeLimit":"One Free cluster per organization with 5 GB storage (~1M 768-dim vectors), 2.5M vCUs/month, and up to 5 collections, no credit card required.","bestFor":"Learning, prototypes, and personal RAG or semantic-search projects that need a managed vector database without standing up Milvus yourself.","caveats":"Free cluster is limited to one per org on a shared GCP environment, capped at 5 collections and 2.5M vCUs/month, with reduced rate limits (2 MB/s writes, search nq \u003c= 10, topK \u003c= 1,024) and no SLA. Exceeding the monthly vCU allowance pauses workloads until reset or upgrade.","sourceUrl":"https://zilliz.com/pricing","highlights":["5 GB storage and 2.5M vCUs/month free, perpetually, with no credit card","Fully managed Milvus, the leading open-source vector database, on GCP","Up to 5 collections with AUTOINDEX, hybrid search, and core APIs","Built-in encryption in transit and at rest, plus console-managed backups"],"tags":["vector-database","milvus","rag","semantic-search","embeddings","ai","managed","gcp"],"pricingModel":"freemium","renewal":"monthly","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"The Free cluster is capped at 5 GB of storage, equivalent to roughly 1 million 768-dimensional vectors per cluster.","label":"5 GB storage per cluster","renewable":false,"sentiment":"neutral","severity":"medium","type":"single_resource_cap","unit":"GB","value":5},{"detail":"Read and write operations consume virtual compute units (vCUs); the Free plan includes 2.5 million vCUs per month, after which workloads are throttled/paused until the monthly reset or an upgrade.","label":"2.5M vCUs per month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"vCUs/mo","value":2500000},{"detail":"A Free cluster allows a maximum of 5 collections (Serverless allows 100).","label":"Up to 5 collections","renewable":false,"sentiment":"neutral","severity":"medium","type":"single_resource_cap","unit":"collections","value":5},{"detail":"Only 1 Free cluster is permitted per organization; you can drop and recreate it but cannot run multiple Free clusters.","label":"One Free cluster per organization","renewable":false,"sentiment":"neutral","severity":"medium","type":"single_resource_cap","unit":"cluster","value":1},{"detail":"Free clusters run only on Google Cloud in a shared (multi-tenant) environment, with no dedicated compute, auto-scaling, replicas, or VPC peering.","label":"GCP-only shared environment","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"Free clusters are limited to 2 MB/s for insert and upsert operations, versus 10 MB/s for Serverless.","label":"2 MB/s insert/upsert rate","renewable":false,"sentiment":"neutral","severity":"low","type":"rate_limit","unit":"MB/s","value":2},{"detail":"On Free clusters search nq must be \u003c= 10 and topK \u003c= 1,024 entities, and a maximum of 2 shards is allowed per collection.","label":"Reduced search limits","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"No uptime SLA is offered on the Free plan; SLAs (99.95%+) require Enterprise or higher paid plans.","label":"No uptime SLA","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"SSO (SAML), granular RBAC, audit logs, private endpoints, CMEK, and HIPAA eligibility are reserved for paid Enterprise/Business Critical plans.","label":"No advanced security/governance","renewable":false,"sentiment":"neutral","severity":"info","type":"feature_restricted"}],"keyMetrics":[{"display":"1 tokens/mo","key":"free_tokens_per_month","label":"Monthly tokens","note":"policy_default: Conservative scoring fallback: no renewable monthly token allowance is published in this item.","primary":true,"unit":"tokens/month","value":1},{"display":"1 tokens/min","key":"tokens_per_minute","label":"Tokens/min","note":"policy_default: Conservative scoring fallback: no comparable token throughput allowance is published in this item.","primary":false,"unit":"tokens/minute","value":1},{"display":"5 GB","key":"storage_gb","label":"Storage","primary":false,"unit":"GB","value":5}],"verified":true,"timestamp":"2026-06-17T00:00:00+00:00","badges":["🟢 Free forever","🟢 Commercial OK","🟠 Single resource","🟢 Embeddings/RAG"],"valueModel":{"bucket":"free","calc":{"allowance_usd_mo":1.5,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":5.2,"unit":"score","value":5.2,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.1,"b":3,"formula":"fixed_log_v1","key":"capacity","label":"Free value","provenance":"provider_quantity_times_reference_price","score":3.4,"unit":"$/mo","value":1.50000002},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Search fit","provenance":"derived_from_limitations_badges_and_access_terms","score":6,"unit":"score","value":6},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Access","provenance":"derived_from_limitations_badges_and_access_terms","score":8.3,"unit":"score","value":8.3}],"required_axis_count":3,"score":5.2,"score_formula":"weighted_components_v1","value":1.5,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"token_month","source":"frontmatter_keyMetrics","unit":"tokens/month","value":1},{"key":"token_minute","source":"frontmatter_keyMetrics","unit":"tokens/minute","value":1},{"key":"db_storage_gb_month","source":"frontmatter_keyMetrics","unit":"GB","value":5}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"embeddings/upstash-vector","slug":"upstash-vector","category":"embeddings","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Upstash | Vector","provider":"Upstash","planName":"freemium","kind":"Managed vector database","description":"Serverless vector database for embeddings, semantic search, metadata filtering, namespaces, and RAG retrieval.","logo":"/logos/upstash.svg","url":"https://upstash.com/docs/vector","freeLimit":"Free tier: 10K daily query/update operations, 1 GB max data/metadata size, 1,536 max dimensions, and 100 namespaces.","bestFor":"Lower-level embedding storage and semantic retrieval when you want to control chunking, embeddings, namespaces, and metadata filters.","caveats":"More primitive than Upstash | Search: you manage document chunking, embedding strategy, and retrieval shape. Use Search when you want a higher-level document search product.","sourceUrl":"https://upstash.com/pricing/vector","highlights":["10K query/update operations per day free","Dense, sparse, and hybrid vector search options","Metadata filtering and namespaces","Good fit for custom RAG indexes"],"tags":["vector-db","embeddings","semantic-search","rag","serverless"],"pricingModel":"freemium","renewal":"monthly","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"Free tier allows 10,000 query and update operations per day, resetting each day.","label":"10K daily operations","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"operations/day","value":10000},{"detail":"The free index is hard-capped at 1 GB of total vector data and metadata storage, with no overflow.","label":"1 GB total data/metadata","renewable":false,"sentiment":"con","severity":"medium","type":"total_volume_cap","unit":"GB","value":1},{"detail":"Only one Vector index is available on the free plan; the FAQ notes up to 10 indexes free but the plan table shows '1 Free DB'.","label":"1 free database index","renewable":false,"sentiment":"neutral","severity":"medium","type":"single_resource_cap","unit":"index","value":1},{"detail":"Free tier caps vector dimensions at 1,536, while paid tiers support up to 3,072 (Pay-as-you-go/Fixed) or 5,000 (Pro).","label":"Max 1,536 dimensions","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"dimensions","value":1536},{"detail":"Free tier is limited to 100 namespaces; paid tiers allow 10,000 or unlimited namespaces.","label":"100 namespaces max","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"namespaces","value":100},{"detail":"The free plan has no uptime SLA; paid plans offer 99.9% or 99.99% SLA.","label":"No uptime SLA","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Free tier receives only community support; email and dedicated Slack/Discord support require paid plans.","label":"Community support only","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"Free indexes scale to zero when idle, causing cold-start latency on first request after inactivity.","label":"Scale to zero / shared infra","renewable":false,"sentiment":"neutral","severity":"medium","type":"performance_throttled"},{"detail":"Free plan is available in N. Virginia (AWS), Ireland (AWS), and Iowa (GCP) only; region choice may not match user's geography.","label":"Limited regions","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"The free plan supports up to 200 million vectors-times-dimensions capacity, versus 2 billion on paid plans.","label":"Max 200M vectors x dimensions","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"vectors x dimensions","value":200000000}],"keyMetrics":[{"display":"1 tokens/mo","key":"free_tokens_per_month","label":"Monthly tokens","note":"policy_default: Conservative scoring fallback: no renewable monthly token allowance is published in this item.","primary":true,"unit":"tokens/month","value":1},{"display":"1 tokens/min","key":"tokens_per_minute","label":"Tokens/min","note":"policy_default: Conservative scoring fallback: no comparable token throughput allowance is published in this item.","primary":false,"unit":"tokens/minute","value":1},{"display":"1 GB","key":"storage_gb","label":"Storage","primary":false,"unit":"GB","value":1}],"verified":true,"timestamp":"2026-06-17T00:00:00+00:00","badges":["🟢 Free forever","🟢 Commercial OK","🔴 One-time","🟠 Single resource","🟢 Embeddings/RAG"],"auditNote":"no FPS in parallel report — used audit recommendation","valueModel":{"bucket":"free","calc":{"allowance_usd_mo":0.3,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.1,"unit":"score","value":4.1,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.1,"b":3,"formula":"fixed_log_v1","key":"capacity","label":"Free value","provenance":"provider_quantity_times_reference_price","score":2,"unit":"$/mo","value":0.30000002},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Search fit","provenance":"derived_from_limitations_badges_and_access_terms","score":5.2,"unit":"score","value":5.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Access","provenance":"derived_from_limitations_badges_and_access_terms","score":7.6,"unit":"score","value":7.6}],"required_axis_count":3,"score":4.1,"score_formula":"weighted_components_v1","value":0.3,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"token_month","source":"frontmatter_keyMetrics","unit":"tokens/month","value":1},{"key":"token_minute","source":"frontmatter_keyMetrics","unit":"tokens/minute","value":1},{"key":"db_storage_gb_month","source":"frontmatter_keyMetrics","unit":"GB","value":1}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"embeddings/qdrant-cloud-free","slug":"qdrant-cloud-free","category":"embeddings","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Qdrant Cloud Free Tier","provider":"Qdrant Cloud Free Tier","planName":"free","kind":"Managed vector database","description":"Managed vector search database with a free-forever single-node cluster for prototypes, including free cloud inference on select models.","logo":"/logos/qdrant-cloud-free.svg","url":"https://qdrant.tech/pricing/","freeLimit":"Free forever: single-node cluster with 0.5 vCPU, 1GB RAM, 4GB disk, plus free cloud inference with selected models.","bestFor":"Embedding storage and semantic/vector search for RAG and AI prototypes without standing up your own Qdrant.","caveats":"Single tiny node (0.5 vCPU / 1GB RAM) with no HA; for testing and prototypes, not production scale.","sourceUrl":"https://qdrant.tech/pricing/","highlights":["Verified free-forever single-node cluster","Includes free cloud inference with selected models","Purpose-built vector search for AI workloads","First-party managed Qdrant"],"tags":["vector-db","embeddings","semantic-search","rag","managed"],"pricingModel":"freemium","renewal":"none","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"Free tier allows exactly one single-node cluster; you cannot create additional clusters without upgrading.","label":"One cluster only","renewable":false,"sentiment":"neutral","severity":"high","type":"single_resource_cap"},{"detail":"Storage is hard-capped at 4 GB disk; exceeding it requires upgrading to the paid Standard Tier.","label":"4 GB disk cap","renewable":false,"sentiment":"con","severity":"high","type":"total_volume_cap","unit":"GB","value":4},{"detail":"Cluster RAM is fixed at 1 GB; data that exceeds RAM capacity will degrade query performance or require an upgrade.","label":"1 GB RAM cap","renewable":false,"sentiment":"con","severity":"high","type":"total_volume_cap","unit":"GB","value":1},{"detail":"Only 0.5 vCPU is allocated, meaning compute is severely limited and shared — not suitable for latency-sensitive production traffic.","label":"0.5 vCPU shared node","renewable":false,"sentiment":"neutral","severity":"medium","type":"performance_throttled","unit":"vCPU","value":0.5},{"detail":"Free tier is a single-node cluster with no multi-AZ setup, no automatic failover, and no uptime SLA.","label":"No high availability","renewable":false,"sentiment":"neutral","severity":"high","type":"feature_restricted"},{"detail":"Automated backups and disaster recovery are absent on the free tier; custom automation would be needed.","label":"No backup & disaster recovery","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Horizontal scale-out (multi-node, shard rebalancing) is not available; the cluster is a single node only.","label":"No horizontal scaling","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"GPU-accelerated indexing is reserved for Standard and Premium tiers.","label":"No GPU indexing","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"Cloud inference is included but restricted to free/selected models; paid inference models require the Standard Tier.","label":"Cloud inference: free models only","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"No SLA-backed support; only community-level best-effort support is available on the free tier.","label":"Community support only","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"}],"keyMetrics":[{"display":"1 tokens/mo","key":"free_tokens_per_month","label":"Monthly tokens","note":"policy_default: Conservative scoring fallback: no renewable monthly token allowance is published in this item.","primary":true,"unit":"tokens/month","value":1},{"display":"1 tokens/min","key":"tokens_per_minute","label":"Tokens/min","note":"policy_default: Conservative scoring fallback: no comparable token throughput allowance is published in this item.","primary":false,"unit":"tokens/minute","value":1},{"display":"4 GB","key":"storage_gb","label":"Storage","primary":false,"unit":"GB","value":4}],"verified":true,"timestamp":"2026-06-16T00:00:00+00:00","badges":["🟢 Free forever","🟢 Commercial OK","🟠 Single resource","🔴 One-time","🟢 Embeddings/RAG"],"valueModel":{"bucket":"free","calc":{"allowance_usd_mo":1.2,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.7,"unit":"score","value":4.7,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.1,"b":3,"formula":"fixed_log_v1","key":"capacity","label":"Free value","provenance":"provider_quantity_times_reference_price","score":3.2,"unit":"$/mo","value":1.20000002},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Search fit","provenance":"derived_from_limitations_badges_and_access_terms","score":5.2,"unit":"score","value":5.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Access","provenance":"derived_from_limitations_badges_and_access_terms","score":7.6,"unit":"score","value":7.6}],"required_axis_count":3,"score":4.7,"score_formula":"weighted_components_v1","value":1.2,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"token_month","source":"frontmatter_keyMetrics","unit":"tokens/month","value":1},{"key":"token_minute","source":"frontmatter_keyMetrics","unit":"tokens/minute","value":1},{"key":"db_storage_gb_month","source":"frontmatter_keyMetrics","unit":"GB","value":4}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"embeddings/jina-embeddings-rerank","slug":"jina-embeddings-rerank","category":"embeddings","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Jina AI Search Foundation","provider":"Jina AI Search Foundation","planName":"freemium","kind":"Hosted embeddings + reranker API (free tokens)","description":"Jina AI provides best-in-class multilingual embeddings and rerankers through a unified Search Foundation API, with free tokens auto-granted to every new key and no credit card required.","logo":"/logos/jina-embeddings-rerank.png","url":"https://jina.ai/embeddings/","freeLimit":"10 million free tokens per new API key (shared across Embeddings, Reranker, Reader, Classifier and DeepSeek/segment APIs), no credit card. Free tier: 100 RPM & 100K TPM for embeddings/rerank. jina-embeddings-v4 is additionally free (throttled).","bestFor":"RAG and search builders who need both embeddings and a reranker from one key with a generous free allowance.","caveats":"10M tokens is a one-time grant shared across all Jina APIs (not monthly-refilling); throughput on the free tier is capped at 100 RPM; some models (v4) are CC-BY-NC / noncom.","sourceUrl":"https://jina.ai/embeddings/","highlights":["Confirmed on official site: 10M free tokens per key, no credit card","Single key covers embeddings, reranker, reader, and classifier","OpenAI-compatible embeddings schema (api.jina.ai/v1/embeddings)","Strong multilingual + multimodal models (v3, v4, clip-v2)"],"tags":["embeddings","reranker","rag","free-tokens","multilingual"],"pricingModel":"freemium","renewal":"none","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"Every new API key receives exactly 10 million free tokens shared across all Jina APIs (Embeddings, Reranker, Reader, Classifier, DeepSearch); once exhausted, paid top-up is required.","label":"10M one-time token grant","renewable":false,"sentiment":"con","severity":"high","type":"total_volume_cap","unit":"tokens","value":10000000},{"detail":"Free API keys are capped at 100 requests per minute and 100,000 tokens per minute on the Embedding and Reranker APIs.","label":"100 RPM / 100K TPM (Embeddings & Reranker)","renewable":true,"sentiment":"neutral","severity":"medium","type":"rate_limit","unit":"requests/minute","value":100},{"detail":"Free keys allow only 2 concurrent in-flight requests across all Jina APIs, limiting parallelism in batch workloads.","label":"2 concurrent requests on free tier","renewable":true,"sentiment":"neutral","severity":"medium","type":"rate_limit","unit":"concurrent requests","value":2},{"detail":"The Classifier train and classify endpoints are limited to 25 RPM and 25,000 TPM on free keys.","label":"25 RPM / 25K TPM (Classifier API)","renewable":true,"sentiment":"neutral","severity":"low","type":"rate_limit","unit":"requests/minute","value":25},{"detail":"Free keys are limited to 50 requests per minute on the DeepSearch endpoint.","label":"50 RPM (DeepSearch)","renewable":true,"sentiment":"neutral","severity":"low","type":"rate_limit","unit":"requests/minute","value":50},{"detail":"The jina-embeddings-v4 model is released under the Qwen Research License (CC BY-NC), which prohibits commercial or production use; only v3 and older models can be used commercially via the API.","label":"jina-embeddings-v4 noncom only","renewable":false,"sentiment":"con","severity":"high","type":"no_commercial_use"},{"detail":"Because Jina cannot monetize v4, they intentionally cap its API throughput to control infrastructure costs; high-volume use is explicitly unsupported and callers are directed to self-host.","label":"jina-embeddings-v4 intentionally throttled","renewable":false,"sentiment":"neutral","severity":"medium","type":"performance_throttled"},{"detail":"Low-traffic models are offloaded during idle periods; the first request after inactivity triggers a warm-up delay of several seconds.","label":"Cold starts on serverless models","renewable":false,"sentiment":"neutral","severity":"low","type":"cold_start_sleep"},{"detail":"The 10M free tokens are shared across Reader, Embeddings, Reranker, Classifier, and DeepSearch — Reader searches alone cost 10,000 tokens per request, so heavy multi-product use can drain the pool quickly.","label":"Token pool shared across all APIs","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"}],"keyMetrics":[{"display":"10M tokens","key":"one_time_tokens","label":"Token grant","note":"one-time grant; kept separate from monthly scoring value","primary":true,"unit":"tokens","value":10000000},{"display":"1 tokens/mo","key":"free_tokens_per_month","label":"Monthly tokens","note":"policy_default: Conservative scoring fallback: no renewable monthly token allowance is published in this item.","primary":false,"unit":"tokens/month","value":1},{"display":"100K tokens/min","key":"tokens_per_minute","label":"Tokens/min","primary":false,"unit":"tokens/minute","value":100000},{"display":"0.001 GB","key":"storage_gb","label":"Storage","note":"policy_default: Conservative scoring fallback: no comparable included storage allowance is published in this item.","primary":false,"unit":"GB","value":0.001},{"display":"No daily cap","key":"requests_per_day","label":"Requests/day","note":"100 RPM","primary":false,"unit":"req/day"},{"display":"Yes","key":"commercial_use","label":"Commercial use","note":"v4 noncom","primary":false,"unit":"yes/no"}],"verified":true,"timestamp":"2026-06-16T00:00:00+00:00","badges":["🟢 Free forever","🔴 Non-commercial","🔴 One-time","🟠 Cold start","🟢 Embeddings/RAG","🟢 Never sleeps"],"bestFreeModel":"jina-embeddings-v5-omni-small","bestFreeIndex":0,"freeModelCount":22,"totalModelCount":22,"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":0,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":2.7,"unit":"score","value":2.7,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.1,"b":3,"formula":"fixed_log_v1","key":"capacity","label":"Free value","provenance":"provider_quantity_times_reference_price","score":1,"unit":"$/mo","value":0.00030001999999999996},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Search fit","provenance":"derived_from_limitations_badges_and_access_terms","score":4.1,"unit":"score","value":4.1},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Access","provenance":"derived_from_limitations_badges_and_access_terms","score":4.8,"unit":"score","value":4.8}],"required_axis_count":3,"score":2.7,"score_formula":"weighted_components_v1","value":0,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"allowed_use":"noncom","source":"stated","value":false},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"token_month","source":"frontmatter_keyMetrics","unit":"tokens/month","value":1},{"key":"token_minute","source":"frontmatter_keyMetrics","unit":"tokens/minute","value":100000},{"key":"db_storage_gb_month","source":"frontmatter_keyMetrics","unit":"GB","value":0.001}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"embeddings/voyage-ai-embeddings","slug":"voyage-ai-embeddings","category":"embeddings","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Voyage AI","provider":"Voyage AI","planName":"finite_credit","kind":"Hosted embeddings + reranker API (free tokens)","description":"Voyage AI (now part of MongoDB) offers state-of-the-art retrieval embeddings and rerankers, with a large free token allowance included on every account.","logo":"/logos/apis/voyage-ai.png","url":"https://docs.voyageai.com/docs/pricing","freeLimit":"First 200 million free tokens for most models (voyage-4-large/4/4-lite, voyage-context-3, voyage-code-3, rerank-2.5, rerank-2.5-lite) per account; 50M for specialized models (finance/law/code-2). No card needed to start.","bestFor":"Production RAG pipelines wanting top-tier embeddings plus a reranker with one of the largest free token allowances available.","caveats":"200M tokens is a one-time per-account allowance (not monthly); rate-limit tiers progress with paid usage; multimodal pixels counted separately.","sourceUrl":"https://docs.voyageai.com/docs/pricing","highlights":["Confirmed on docs: 200M free tokens for most embedding and reranker models","Includes rerank-2.5 and rerank-2.5-lite rerankers in the free tier","Code- and domain-specialized models (voyage-code-3, finance, law)","Backed by MongoDB; strong retrieval benchmark performance"],"tags":["embeddings","reranker","rag","free-tokens","mongodb"],"pricingModel":"finite_credit","renewal":"none","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"voyage-4-large, voyage-4, voyage-4-lite, voyage-context-3, voyage-code-3, rerank-2.5, and rerank-2.5-lite each receive 200 million free tokens per account — a one-time lifetime allowance that does not reset.","label":"200M free tokens (main models)","renewable":false,"sentiment":"con","severity":"medium","type":"total_volume_cap","unit":"tokens","value":200000000},{"detail":"Older specialized models voyage-multilingual-2, voyage-finance-2, voyage-law-2, and voyage-code-2 receive only 50 million free tokens per account.","label":"50M free tokens (legacy/specialized models)","renewable":false,"sentiment":"con","severity":"medium","type":"total_volume_cap","unit":"tokens","value":50000000},{"detail":"voyage-multimodal-3.5 and voyage-multimodal-3 receive 200M free text tokens and 150B free pixels per account as a one-time allowance.","label":"200M tokens + 150B pixels (multimodal)","renewable":false,"sentiment":"con","severity":"medium","type":"total_volume_cap","unit":"tokens","value":200000000},{"detail":"Free-tier accounts start at the lowest RPM/TPM rate-limit tier; limits scale up only as paid usage accumulates.","label":"Lowest rate-limit tier on free usage","renewable":true,"sentiment":"neutral","severity":"medium","type":"rate_limit"},{"detail":"Models predating the current lineup (voyage-3, voyage-3-large, voyage-3.5, rerank-1, etc.) have no free token allocation and are billed from the first token.","label":"No free tokens on older models","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"Fine-tuned dedicated instances and support channels require a sales engagement; there is no self-serve free-tier path.","label":"No fine-tuned model access","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"}],"keyMetrics":[{"display":"200M tokens","key":"one_time_tokens","label":"Token grant","note":"one-time grant; kept separate from monthly scoring value","primary":true,"unit":"tokens","value":200000000},{"display":"1 tokens/mo","key":"free_tokens_per_month","label":"Monthly tokens","note":"policy_default: Conservative scoring fallback: no renewable monthly token allowance is published in this item.","primary":false,"unit":"tokens/month","value":1},{"display":"8M tokens/min","key":"tokens_per_minute","label":"Tokens/min","primary":false,"unit":"tokens/minute","value":8000000},{"display":"0.001 GB","key":"storage_gb","label":"Storage","note":"policy_default: Conservative scoring fallback: no comparable included storage allowance is published in this item.","primary":false,"unit":"GB","value":0.001},{"display":"No daily cap","key":"requests_per_day","label":"Requests/day","note":"2,000 RPM","primary":false,"unit":"req/day"},{"display":"Yes","key":"commercial_use","label":"Commercial use","primary":false,"unit":"yes/no"}],"verified":true,"timestamp":"2026-06-16T00:00:00+00:00","badges":["🟢 Free forever","🟢 Commercial OK","🔴 One-time","🟢 Embeddings/RAG"],"bestFreeModel":"voyage-4-large","bestFreeIndex":0,"freeModelCount":12,"totalModelCount":25,"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":0,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":3.6,"unit":"score","value":3.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":1,"b":1,"formula":"fixed_log_v1","key":"capacity","label":"Free value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"$ one-time","value":1},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Search fit","provenance":"derived_from_limitations_badges_and_access_terms","score":5.7,"unit":"score","value":5.7},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Access","provenance":"derived_from_limitations_badges_and_access_terms","score":6.8,"unit":"score","value":6.8}],"required_axis_count":3,"score":3.6,"score_formula":"weighted_components_v1","value":1,"value_kind":"one_time_usd","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"stated","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"token_month","source":"frontmatter_keyMetrics","unit":"tokens/month","value":1},{"key":"token_minute","source":"frontmatter_keyMetrics","unit":"tokens/minute","value":8000000},{"key":"db_storage_gb_month","source":"frontmatter_keyMetrics","unit":"GB","value":0.001}],"raw":{"pricingModel":"finite_credit"},"subtop":"renewable"}},{"id":"embeddings/weaviate-cloud","slug":"weaviate-cloud","category":"embeddings","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Weaviate Cloud","provider":"Weaviate Cloud","planName":"freemium","kind":"Managed Vector Database","description":"Fully managed, AI-native vector database with a forever-free Sandbox cluster for building search and RAG apps.","logo":"/logos/weaviate-cloud-serverless.png","url":"https://weaviate.io","freeLimit":"One always-free Sandbox cluster per user with 100,000 objects, 1 GB memory, 10 GB disk, 1 collection (up to 3 tenants), plus 2,000 embedding requests/day and 1,000 Query Agent requests/month.","bestFor":"Prototyping vector search, RAG, and AI agent apps without a credit card or self-hosting overhead.","caveats":"Hard cap of 100,000 objects and a single collection limits it to prototypes and small datasets; only 1 free cluster per user, and embedding/Query Agent calls are rate-limited. Some third-party sources still reference an older 14-day trial sandbox, but the official pricing page now lists the Free tier as always free.","sourceUrl":"https://weaviate.io/pricing","highlights":["Forever-free Sandbox cluster, no credit card required","Bundled inference: 2,000 embedding requests/day plus 1,000 Query Agent requests/month","Fully managed vector DB with hybrid (vector + keyword) search and built-in vectorizers","One-click upgrade to paid Serverless when you outgrow the free limits"],"tags":["vector-database","ai","rag","semantic-search","managed","embeddings","always-free"],"pricingModel":"freemium","renewal":"none","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"The free Sandbox is limited to a single cluster per user account; you cannot run multiple free clusters.","label":"1 cluster per user","renewable":false,"sentiment":"neutral","severity":"medium","type":"single_resource_cap","unit":"cluster","value":1},{"detail":"The free Sandbox stores at most 100,000 objects (vectors); exceeding this requires upgrading to a paid Serverless plan.","label":"100,000 object cap","renewable":false,"sentiment":"con","severity":"high","type":"total_volume_cap","unit":"objects","value":100000},{"detail":"Fixed Sandbox sizing of 10 GB disk and 1 GB memory; not scalable on the free tier.","label":"10 GB disk / 1 GB memory","renewable":false,"sentiment":"neutral","severity":"medium","type":"single_resource_cap","unit":"GB disk","value":10},{"detail":"Free Sandbox allows only a single collection (schema class) with up to 3 multi-tenancy tenants.","label":"1 collection, up to 3 tenants","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted","unit":"collection","value":1},{"detail":"Bundled embedding (vectorizer) inference is capped at 2,000 requests per day on the free tier.","label":"2,000 embedding requests/day","renewable":true,"sentiment":"neutral","severity":"medium","type":"rate_limit","unit":"requests/day","value":2000},{"detail":"Bundled Query Agent (agentic query) usage is capped at 1,000 requests per month on the free tier.","label":"1,000 Query Agent requests/month","renewable":true,"sentiment":"neutral","severity":"medium","type":"rate_limit","unit":"requests/month","value":1000},{"detail":"Free tier support is limited to the Weaviate Slack and community forum; no SLA or dedicated support.","label":"Community support only","renewable":false,"sentiment":"neutral","severity":"info","type":"feature_restricted"}],"keyMetrics":[{"display":"1 tokens/mo","key":"free_tokens_per_month","label":"Monthly tokens","note":"policy_default: Conservative scoring fallback: no renewable monthly token allowance is published in this item.","primary":true,"unit":"tokens/month","value":1},{"display":"1 tokens/min","key":"tokens_per_minute","label":"Tokens/min","note":"policy_default: Conservative scoring fallback: no comparable token throughput allowance is published in this item.","primary":false,"unit":"tokens/minute","value":1},{"display":"10 GB","key":"storage_gb","label":"Storage","primary":false,"unit":"GB","value":10}],"verified":true,"timestamp":"2026-06-17T00:00:00+00:00","badges":["🟢 Free forever","🟢 Commercial OK","🟠 Single resource","🔴 One-time","🟢 Embeddings/RAG"],"valueModel":{"bucket":"free","calc":{"allowance_usd_mo":3,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":5.1,"unit":"score","value":5.1,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.1,"b":3,"formula":"fixed_log_v1","key":"capacity","label":"Free value","provenance":"provider_quantity_times_reference_price","score":4,"unit":"$/mo","value":3.00000002},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Search fit","provenance":"derived_from_limitations_badges_and_access_terms","score":5.2,"unit":"score","value":5.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Access","provenance":"derived_from_limitations_badges_and_access_terms","score":7.6,"unit":"score","value":7.6}],"required_axis_count":3,"score":5.1,"score_formula":"weighted_components_v1","value":3,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"token_month","source":"frontmatter_keyMetrics","unit":"tokens/month","value":1},{"key":"token_minute","source":"frontmatter_keyMetrics","unit":"tokens/minute","value":1},{"key":"db_storage_gb_month","source":"frontmatter_keyMetrics","unit":"GB","value":10}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"embeddings/pinecone-builder","slug":"pinecone-builder","category":"embeddings","access":"conditional","accessBucket":"paid","tier":"paid","name":"Pinecone","provider":"Pinecone","planName":"paid","kind":"Paid plan","description":"Builder: $20/mo flat — increased usage limits, up to 10 GB storage, up to 5M write units & 2M read units/mo, 10 indexes, choose cloud/region, Prometheus/Datadog, free support.","logo":"/logos/pinecone-builder.png","url":"https://www.pinecone.io/pricing/","freeLimit":"Paid — $20/mo flat (Builder). Builder: $20/mo flat — increased usage limits, up to 10 GB storage, up to 5M write units & 2M read units/mo, 10 indexes, choose cloud/region, Prometheus/Datadog, free support.","bestFor":"Flat $20 production vector DB with no per-vector metering at this tier — best fixed-price vector entry","caveats":"","tags":["paid","value","embeddings","vector-database"],"pricingModel":"paid","keyMetrics":[{"display":"1 tokens/mo","key":"free_tokens_per_month","label":"Monthly tokens","note":"policy_default: Conservative scoring fallback: no renewable monthly token allowance is published in this item.","primary":true,"unit":"tokens/month","value":1},{"display":"1 tokens/min","key":"tokens_per_minute","label":"Tokens/min","note":"policy_default: Conservative scoring fallback: no comparable token throughput allowance is published in this item.","primary":false,"unit":"tokens/minute","value":1},{"display":"10 GB","key":"storage_gb","label":"Storage","primary":false,"unit":"GB","value":10},{"display":"$20/mo","key":"m0","label":"","primary":false},{"display":"10 GB","key":"m1","label":"","primary":false},{"display":"10 indexes","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Embeddings/RAG"],"valueModel":{"bucket":"paid","calc":{"allowance_usd_mo":3,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":5.8,"unit":"score","value":5.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Free value","provenance":"provider_quantity_times_reference_price","score":3.4,"unit":"×","value":0.150000001},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Search fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Access","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":5.8,"score_formula":"weighted_components_v1","value":0.15,"value_kind":"ratio","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"token_month","source":"frontmatter_keyMetrics","unit":"tokens/month","value":1},{"key":"token_minute","source":"frontmatter_keyMetrics","unit":"tokens/minute","value":1},{"key":"db_storage_gb_month","source":"frontmatter_keyMetrics","unit":"GB","value":10}],"raw":{"offered_price_usd":20,"pricingModel":"paid"},"subtop":"renewable"}},{"id":"embeddings/weaviate-cloud-serverless","slug":"weaviate-cloud-serverless","category":"embeddings","access":"conditional","accessBucket":"free","name":"Weaviate Cloud","provider":"Weaviate Cloud","planName":"freemium","kind":"Paid plan","description":"Serverless: ~$25/mo minimum, then $0.095 per 1M vector dimensions/month (storage) — among the cheapest per-dimension vector rates; free tier (100K objects, 1 GB) to start.","logo":"/logos/weaviate-cloud-serverless.png","url":"https://weaviate.io/pricing","freeLimit":"Free tier (100K objects, 1 GB) to start; Serverless then ~$25/mo minimum, then $0.095 per 1M vector dimensions/month (storage) — among the cheapest per-dimension vector rates.","bestFor":"$0.095 per 1M dimensions — lowest published per-dimension vector storage rate here; dimension-based pricing rewards compression","caveats":"","tags":["paid","value","embeddings","vector-database"],"pricingModel":"freemium","keyMetrics":[{"display":"1 tokens/mo","key":"free_tokens_per_month","label":"Monthly tokens","note":"policy_default: Conservative scoring fallback: no renewable monthly token allowance is published in this item.","primary":true,"unit":"tokens/month","value":1},{"display":"1 tokens/min","key":"tokens_per_minute","label":"Tokens/min","note":"policy_default: Conservative scoring fallback: no comparable token throughput allowance is published in this item.","primary":false,"unit":"tokens/minute","value":1},{"display":"1 GB","key":"storage_gb","label":"Storage","primary":false,"unit":"GB","value":1},{"display":"$25/mo min","key":"m0","label":"","primary":false},{"display":"$0.095/M dim","key":"m1","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Embeddings/RAG"],"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":0.3,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":5.2,"unit":"score","value":5.2,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.1,"b":3,"formula":"fixed_log_v1","key":"capacity","label":"Free value","provenance":"provider_quantity_times_reference_price","score":2,"unit":"$/mo","value":0.30000002},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Search fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Access","provenance":"derived_from_limitations_badges_and_access_terms","score":8.8,"unit":"score","value":8.8}],"required_axis_count":3,"score":5.2,"score_formula":"weighted_components_v1","value":0.3,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"token_month","source":"frontmatter_keyMetrics","unit":"tokens/month","value":1},{"key":"token_minute","source":"frontmatter_keyMetrics","unit":"tokens/minute","value":1},{"key":"db_storage_gb_month","source":"frontmatter_keyMetrics","unit":"GB","value":1}],"raw":{"offered_price_usd":25,"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"embeddings/zilliz-cloud-serverless","slug":"zilliz-cloud-serverless","category":"embeddings","access":"conditional","accessBucket":"free","tier":"paid","name":"Zilliz Cloud","provider":"Zilliz Cloud","planName":"paid","kind":"Paid plan","description":"Serverless: $4 per million vCUs (read+write), storage $0.025-0.30/GB-mo; free tier 5 GB + 2.5M vCUs + 5 collections. Managed Milvus; dedicated from $99/GB/mo if needed.","logo":"/logos/zilliz-cloud-serverless.png","url":"https://zilliz.com/pricing","freeLimit":"Paid — pay-as-you-go from $0 (then $4/M vCUs). Serverless: $4 per million vCUs (read+write), storage $0.025-0.30/GB-mo; free tier 5 GB + 2.5M vCUs + 5 collections. Managed Milvus; dedicated from $99/GB/mo if needed.","bestFor":"Up to ~50x cost cut vs provisioned via hot/cold tiering; true scale-to-zero serverless vectors — fully-managed Milvus by its creators","caveats":"","sourceUrl":"https://zilliz.com/pricing","tags":["paid","value","embeddings","vector-database"],"pricingModel":"freemium","renewal":"monthly","requiresCard":false,"eligibility":["anyone"],"limitations":[{"detail":"Zilliz Cloud Serverless starts from $0 with a free allowance; usage beyond that is $4 per million vCUs plus storage ($0.025-0.30/GB-mo).","label":"Free serverless starting tier","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable"},{"detail":"Serverless collections run on shared, autoscaled capacity rather than dedicated compute.","label":"Shared serverless capacity","renewable":false,"sentiment":"neutral","severity":"low","type":"performance_throttled"}],"keyMetrics":[{"display":"1 tokens/mo","key":"free_tokens_per_month","label":"Monthly tokens","note":"policy_default: Conservative scoring fallback: no renewable monthly token allowance is published in this item.","primary":true,"unit":"tokens/month","value":1},{"display":"1 tokens/min","key":"tokens_per_minute","label":"Tokens/min","note":"policy_default: Conservative scoring fallback: no comparable token throughput allowance is published in this item.","primary":false,"unit":"tokens/minute","value":1},{"display":"5 GB","key":"storage_gb","label":"Storage","primary":false,"unit":"GB","value":5},{"display":"from $0","key":"m0","label":"","primary":false},{"display":"$4/M vCU","key":"m1","label":"","primary":false},{"display":"5 GB free","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","badges":["🟢 Commercial OK","🟢 Embeddings/RAG"],"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":1.5,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":5.6,"unit":"score","value":5.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.1,"b":3,"formula":"fixed_log_v1","key":"capacity","label":"Free value","provenance":"provider_quantity_times_reference_price","score":3.4,"unit":"$/mo","value":1.50000002},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Search fit","provenance":"derived_from_limitations_badges_and_access_terms","score":7,"unit":"score","value":7},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Access","provenance":"derived_from_limitations_badges_and_access_terms","score":8.8,"unit":"score","value":8.8}],"required_axis_count":3,"score":5.6,"score_formula":"weighted_components_v1","value":1.5,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"token_month","source":"frontmatter_keyMetrics","unit":"tokens/month","value":1},{"key":"token_minute","source":"frontmatter_keyMetrics","unit":"tokens/minute","value":1},{"key":"db_storage_gb_month","source":"frontmatter_keyMetrics","unit":"GB","value":5}],"raw":{"offered_price_usd":99,"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"embeddings/qdrant-cloud-standard","slug":"qdrant-cloud-standard","category":"embeddings","access":"conditional","accessBucket":"paid","tier":"paid","name":"Qdrant Cloud","provider":"Qdrant Cloud","planName":"paid","kind":"Paid plan","description":"Standard: dedicated cluster, usage-based (~$0.078/GB-hour RAM); a small ~1 GB cluster runs ~$25-30/mo. 99.5% SLA, vertical/horizontal scaling, backups, free inference tokens.","logo":"/logos/qdrant-cloud-standard.png","url":"https://qdrant.tech/pricing/","freeLimit":"Paid — from ~$25/mo (Standard, usage-based). Standard: dedicated cluster, usage-based (~$0.078/GB-hour RAM); a small ~1 GB cluster runs ~$25-30/mo. 99.5% SLA, vertical/horizontal scaling, backups, free inference tokens.","bestFor":"Among the cheapest dedicated vector clusters; ~$0.095/GB storage class economics at scale — high-performance Rust vector DB, strong price/recall with ","caveats":"","tags":["paid","value","embeddings","vector-database"],"pricingModel":"paid","keyMetrics":[{"display":"1 tokens/mo","key":"free_tokens_per_month","label":"Monthly tokens","note":"policy_default: Conservative scoring fallback: no renewable monthly token allowance is published in this item.","primary":true,"unit":"tokens/month","value":1},{"display":"1 tokens/min","key":"tokens_per_minute","label":"Tokens/min","note":"policy_default: Conservative scoring fallback: no comparable token throughput allowance is published in this item.","primary":false,"unit":"tokens/minute","value":1},{"display":"0.001 GB","key":"storage_gb","label":"Storage","note":"policy_default: Conservative scoring fallback: no comparable included storage allowance is published in this item.","primary":false,"unit":"GB","value":0.001},{"display":"~$25/mo","key":"m0","label":"","primary":false},{"display":"$0.078/GBh","key":"m1","label":"","primary":false},{"display":"99.5% SLA","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Embeddings/RAG"],"valueModel":{"bucket":"paid","calc":{"allowance_usd_mo":0,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Free value","provenance":"provider_quantity_times_reference_price","score":1,"unit":"×","value":0.000012000799999999999},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Search fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Access","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0,"value_kind":"ratio","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"token_month","source":"frontmatter_keyMetrics","unit":"tokens/month","value":1},{"key":"token_minute","source":"frontmatter_keyMetrics","unit":"tokens/minute","value":1},{"key":"db_storage_gb_month","source":"frontmatter_keyMetrics","unit":"GB","value":0.001}],"raw":{"offered_price_usd":25,"pricingModel":"paid"},"subtop":"renewable"}}],"categories":[{"id":"llm-providers","label":"LLM","description":"Free or BYOK model APIs"},{"id":"embeddings","label":"Embeddings","description":"Embedding, reranker, and vector search services"},{"id":"databases","label":"DB","description":"Structured app state"},{"id":"app-platforms","label":"Apps","description":"Deploy full apps & services"},{"id":"agent-deployment","label":"Agents","description":"Deploy & run AI agents"},{"id":"serverless-functions","label":"Functions","description":"Run functions, edge & jobs"},{"id":"schedulers","label":"Schedulers","description":"Cron jobs & recurring tasks"},{"id":"storage","label":"Storage","description":"Files, datasets, bundles"},{"id":"workspaces","label":"Workspace","description":"Notion-like user surfaces"},{"id":"domains","label":"Domains","description":"Free hostnames and DNS"}],"selectedCategory":"embeddings","showSidebar":true,"categoryCounts":{"agent-deployment":88,"app-platforms":55,"databases":61,"domains":64,"embeddings":11,"llm-providers":105,"schedulers":50,"serverless-functions":58,"storage":56,"workspaces":70},"softwareCategoryCounts":{"agent-deployment":40,"agent-frameworks":39,"agents":41,"browser-computer-use":28,"chat-ui":27,"data-extraction":31,"fine-tuning":36,"guardrails-security":27,"inference-serving":33,"llm-gateways":37,"mcp-tools":30,"observability-evals":36,"orchestration":20,"rag-vector-memory":39,"sandboxes":32,"voice-realtime":41,"web-crawl-search":40},"apiCategoryCounts":{"ai-ml":34,"business":31,"communication":37,"data-open":24,"dev-tools":31,"entertainment":37,"finance-crypto":35,"food-travel":26,"geo-maps":36,"gov-science":27,"media-images":31,"news-media":32,"search-web":35,"social-comm":26,"sports":34,"weather":30},"agentToolCategoryCounts":{"mcp-cloud":26,"mcp-data":32,"mcp-dev":33,"mcp-other":10,"mcp-productivity":33,"mcp-search":34,"plugins":19,"prompt-libraries":24,"registries":16,"resources":25,"skills":26,"toolkits":22},"totalOfferCount":2001,"documentationSections":[{"id":"methodology","name":"Methodology","items":[{"id":"generosity-measurement","title":"How ratings are computed","href":"/documentation/generosity-measurement"},{"id":"free-tier-conditions","title":"Free-Tier Conditions & Limitations","href":"/documentation/free-tier-conditions"},{"id":"badges-and-labels","title":"Badges, labels & markers","href":"/documentation/badges-and-labels"}]}],"recipeSections":[{"id":"personal-assistant","name":"Personal Assistant","items":[{"id":"archive/notion-upstash-telegram-morning-agent","title":"Morning Notion Task Digest to Telegram","href":"/recipes/archive/notion-upstash-telegram-morning-agent"},{"id":"archive/theme0-langgraph-gmail-email-assistant","title":"Human-in-the-Loop Gmail Email Assistant","href":"/recipes/archive/theme0-langgraph-gmail-email-assistant"},{"id":"archive/theme0-google-calendar-mcp-agent","title":"Natural-Language Google Calendar Agent (MCP)","href":"/recipes/archive/theme0-google-calendar-mcp-agent"},{"id":"archive/theme0-second-brain-notes-rag","title":"Second-Brain Notes Assistant (RAG over your vault)","href":"/recipes/archive/theme0-second-brain-notes-rag"},{"id":"archive/theme0-crewai-daily-briefing-crew","title":"Multi-Agent Daily Briefing Crew","href":"/recipes/archive/theme0-crewai-daily-briefing-crew"}]},{"id":"scheduled-automation","name":"Scheduled Automation","items":[{"id":"archive/theme6-scheduled-log-anomaly-sentinel","title":"Scheduled Log-Anomaly Sentinel with Vector Search","href":"/recipes/archive/theme6-scheduled-log-anomaly-sentinel"},{"id":"archive/theme5-triggerdev-youtube-content-digest","title":"Scheduled YouTube-to-Email Content Digest","href":"/recipes/archive/theme5-triggerdev-youtube-content-digest"},{"id":"archive/theme5-firecrawl-competitor-newsletter","title":"Curated Competitor-News Newsletter from Live Crawls","href":"/recipes/archive/theme5-firecrawl-competitor-newsletter"},{"id":"archive/theme3-trigger-dev-hacker-news-scrape-summary","title":"Scheduled Hacker News Scrape, Summarize, and Email Digest","href":"/recipes/archive/theme3-trigger-dev-hacker-news-scrape-summary"},{"id":"archive/theme3-firecrawl-crawl-docs-upstash-vector-rag","title":"Crawl a Documentation Site into an Upstash Vector RAG Index","href":"/recipes/archive/theme3-firecrawl-crawl-docs-upstash-vector-rag"},{"id":"archive/theme3-langchain-incremental-indexing-pgvector-sync","title":"Incrementally Sync a Source into a pgvector Index Without Re-Embedding","href":"/recipes/archive/theme3-langchain-incremental-indexing-pgvector-sync"},{"id":"archive/theme0-firecrawl-news-digest-email","title":"Daily Web Digest to Your Inbox (Firecrawl)","href":"/recipes/archive/theme0-firecrawl-news-digest-email"},{"id":"archive/theme0-triggerdev-per-user-morning-briefing","title":"Per-User Morning Briefing on Trigger.dev","href":"/recipes/archive/theme0-triggerdev-per-user-morning-briefing"},{"id":"archive/theme1-triggerdev-scheduled-scrape-summary-digest","title":"Scheduled Scrape-and-Summarize Digest with Trigger.dev","href":"/recipes/archive/theme1-triggerdev-scheduled-scrape-summary-digest"}]},{"id":"event-automation","name":"Event Automation","items":[{"id":"archive/theme6-autonomous-incident-triage-agent","title":"Autonomous Incident Triage Agent for On-Call SREs","href":"/recipes/archive/theme6-autonomous-incident-triage-agent"},{"id":"archive/theme6-terraform-pr-review-agent-mcp","title":"Terraform PR Agent with the HashiCorp MCP Server","href":"/recipes/archive/theme6-terraform-pr-review-agent-mcp"},{"id":"archive/theme2-langgraph-pr-review-agent","title":"Automated PR Review Agent with LangGraph","href":"/recipes/archive/theme2-langgraph-pr-review-agent"},{"id":"archive/theme5-langgraph-social-media-post-agent","title":"URL to Twitter & LinkedIn Posts with Human Review","href":"/recipes/archive/theme5-langgraph-social-media-post-agent"},{"id":"archive/theme4-crewai-multi-agent-ticket-triage","title":"Multi-Agent Support Ticket Triage with CrewAI","href":"/recipes/archive/theme4-crewai-multi-agent-ticket-triage"},{"id":"archive/theme4-zendesk-webhook-triage-agent","title":"Zendesk Ticket Triage Agent: Classify, Search the KB, Route to Slack","href":"/recipes/archive/theme4-zendesk-webhook-triage-agent"},{"id":"archive/theme3-langchain-invoice-email-to-postgres","title":"Extract Invoice and Receipt Data from Email into Postgres","href":"/recipes/archive/theme3-langchain-invoice-email-to-postgres"},{"id":"archive/theme2-trigger-dev-github-issue-triage","title":"GitHub Issue Triage and PR Review on Trigger.dev","href":"/recipes/archive/theme2-trigger-dev-github-issue-triage"},{"id":"archive/theme1-firecrawl-monitor-competitor-watch-agent","title":"Competitor Web-Change Watch Agent with Firecrawl Monitor","href":"/recipes/archive/theme1-firecrawl-monitor-competitor-watch-agent"}]},{"id":"worker","name":"Worker","items":[{"id":"cloudflare-free-app-stack","title":"Cloudflare Free App Stack","href":"/recipes/cloudflare-free-app-stack"},{"id":"cloudflare-ai-gateway-free-llm-fallback","title":"Cloudflare AI Gateway Free LLM Fallback Chain","href":"/recipes/cloudflare-ai-gateway-free-llm-fallback"},{"id":"archive/theme6-log-clustering-triage-agent","title":"Log Clustering and Triage Agent over MCP","href":"/recipes/archive/theme6-log-clustering-triage-agent"},{"id":"vercel-agent-ai-gateway-central-hub","title":"Vercel Agent AI Gateway Central Hub","href":"/recipes/vercel-agent-ai-gateway-central-hub"},{"id":"archive/theme7-orchestrator-workers-fact-checker","title":"Orchestrator-Workers News Fact-Checker","href":"/recipes/archive/theme7-orchestrator-workers-fact-checker"},{"id":"archive/theme4-firecrawl-docs-to-knowledge-base-agent","title":"Turn a Documentation Site into a Support Agent with Firecrawl and LangGraph","href":"/recipes/archive/theme4-firecrawl-docs-to-knowledge-base-agent"},{"id":"archive/theme5-triggerdev-youtube-content-digest","title":"Scheduled YouTube-to-Email Content Digest","href":"/recipes/archive/theme5-triggerdev-youtube-content-digest"},{"id":"archive/theme7-crewai-hierarchical-content-crew","title":"CrewAI Hierarchical Content Crew","href":"/recipes/archive/theme7-crewai-hierarchical-content-crew"},{"id":"archive/theme5-firecrawl-competitor-newsletter","title":"Curated Competitor-News Newsletter from Live Crawls","href":"/recipes/archive/theme5-firecrawl-competitor-newsletter"},{"id":"archive/theme7-parallel-subagent-research-swarm","title":"Parallel Subagent Research Swarm","href":"/recipes/archive/theme7-parallel-subagent-research-swarm"},{"id":"archive/theme5-content-repurposing-worker","title":"Repurpose One Blog Post into a Week of Channels","href":"/recipes/archive/theme5-content-repurposing-worker"},{"id":"archive/theme2-cover-agent-test-generation","title":"Coverage-Driven Unit Test Generation with Cover-Agent","href":"/recipes/archive/theme2-cover-agent-test-generation"},{"id":"archive/theme3-trigger-dev-hacker-news-scrape-summary","title":"Scheduled Hacker News Scrape, Summarize, and Email Digest","href":"/recipes/archive/theme3-trigger-dev-hacker-news-scrape-summary"},{"id":"archive/theme3-crewai-multi-agent-lead-scraper","title":"Multi-Agent Lead Scraper that Outputs a Structured Leads Table","href":"/recipes/archive/theme3-crewai-multi-agent-lead-scraper"},{"id":"archive/theme3-langchain-invoice-email-to-postgres","title":"Extract Invoice and Receipt Data from Email into Postgres","href":"/recipes/archive/theme3-langchain-invoice-email-to-postgres"},{"id":"archive/theme0-firecrawl-news-digest-email","title":"Daily Web Digest to Your Inbox (Firecrawl)","href":"/recipes/archive/theme0-firecrawl-news-digest-email"},{"id":"archive/theme1-crewai-competitive-analysis-crew","title":"Competitive Analysis Crew that Outputs a Strategic Report","href":"/recipes/archive/theme1-crewai-competitive-analysis-crew"},{"id":"archive/theme1-triggerdev-scheduled-scrape-summary-digest","title":"Scheduled Scrape-and-Summarize Digest with Trigger.dev","href":"/recipes/archive/theme1-triggerdev-scheduled-scrape-summary-digest"}]},{"id":"workflow","name":"Workflow","items":[{"id":"cloudflare-free-app-stack","title":"Cloudflare Free App Stack","href":"/recipes/cloudflare-free-app-stack"},{"id":"cloudflare-ai-gateway-free-llm-fallback","title":"Cloudflare AI Gateway Free LLM Fallback Chain","href":"/recipes/cloudflare-ai-gateway-free-llm-fallback"},{"id":"vercel-agent-ai-gateway-central-hub","title":"Vercel Agent AI Gateway Central Hub","href":"/recipes/vercel-agent-ai-gateway-central-hub"},{"id":"archive/theme6-natural-language-iac-generator","title":"Natural-Language Infrastructure-as-Code Generator","href":"/recipes/archive/theme6-natural-language-iac-generator"},{"id":"archive/theme6-terraform-pr-review-agent-mcp","title":"Terraform PR Agent with the HashiCorp MCP Server","href":"/recipes/archive/theme6-terraform-pr-review-agent-mcp"},{"id":"archive/theme2-langgraph-pr-review-agent","title":"Automated PR Review Agent with LangGraph","href":"/recipes/archive/theme2-langgraph-pr-review-agent"},{"id":"archive/theme7-orchestrator-workers-fact-checker","title":"Orchestrator-Workers News Fact-Checker","href":"/recipes/archive/theme7-orchestrator-workers-fact-checker"},{"id":"archive/theme7-evaluator-optimizer-translation-loop","title":"Evaluator-Optimizer Translation Loop","href":"/recipes/archive/theme7-evaluator-optimizer-translation-loop"},{"id":"archive/theme5-crewai-seo-blog-editorial-crew","title":"Multi-Agent SEO Blog Pipeline that Publishes to Ghost","href":"/recipes/archive/theme5-crewai-seo-blog-editorial-crew"},{"id":"archive/theme7-langgraph-supervisor-research-desk","title":"LangGraph Supervisor Research Desk","href":"/recipes/archive/theme7-langgraph-supervisor-research-desk"},{"id":"archive/theme4-crewai-multi-agent-ticket-triage","title":"Multi-Agent Support Ticket Triage with CrewAI","href":"/recipes/archive/theme4-crewai-multi-agent-ticket-triage"},{"id":"archive/theme5-qdrant-brand-aware-content-rag","title":"Brand-Aware Content Generator with RAG","href":"/recipes/archive/theme5-qdrant-brand-aware-content-rag"},{"id":"archive/theme4-inngest-durable-rag-answer-workflow","title":"Durable RAG Answer Workflow with Inngest","href":"/recipes/archive/theme4-inngest-durable-rag-answer-workflow"},{"id":"archive/theme7-llamaindex-agentworkflow-report-swarm","title":"LlamaIndex AgentWorkflow Report Swarm","href":"/recipes/archive/theme7-llamaindex-agentworkflow-report-swarm"},{"id":"archive/theme5-content-repurposing-worker","title":"Repurpose One Blog Post into a Week of Channels","href":"/recipes/archive/theme5-content-repurposing-worker"},{"id":"archive/theme2-cover-agent-test-generation","title":"Coverage-Driven Unit Test Generation with Cover-Agent","href":"/recipes/archive/theme2-cover-agent-test-generation"},{"id":"archive/theme3-llamaparse-pdf-manuals-qdrant-hybrid-search","title":"Parse Complex PDF Manuals into a Qdrant Hybrid-Search Knowledge Base","href":"/recipes/archive/theme3-llamaparse-pdf-manuals-qdrant-hybrid-search"},{"id":"archive/theme2-openhands-dependency-migration-agent","title":"Autonomous Dependency Migration Agent with OpenHands","href":"/recipes/archive/theme2-openhands-dependency-migration-agent"},{"id":"archive/theme0-crewai-daily-briefing-crew","title":"Multi-Agent Daily Briefing Crew","href":"/recipes/archive/theme0-crewai-daily-briefing-crew"},{"id":"archive/theme2-trigger-dev-github-issue-triage","title":"GitHub Issue Triage and PR Review on Trigger.dev","href":"/recipes/archive/theme2-trigger-dev-github-issue-triage"},{"id":"archive/theme1-firecrawl-deep-research-agent","title":"Deep Research Agent with Firecrawl and an Agent SDK","href":"/recipes/archive/theme1-firecrawl-deep-research-agent"},{"id":"archive/theme2-firecrawl-qdrant-migration-assistant","title":"Live-Docs Migration Assistant with Firecrawl and Qdrant","href":"/recipes/archive/theme2-firecrawl-qdrant-migration-assistant"},{"id":"archive/theme0-triggerdev-per-user-morning-briefing","title":"Per-User Morning Briefing on Trigger.dev","href":"/recipes/archive/theme0-triggerdev-per-user-morning-briefing"},{"id":"archive/theme1-langgraph-exa-deep-research-pipeline","title":"Auditable Deep Research Pipeline with LangGraph and Exa","href":"/recipes/archive/theme1-langgraph-exa-deep-research-pipeline"}]},{"id":"agent-runtime","name":"Agent Runtime","items":[{"id":"cloudflare-free-app-stack","title":"Cloudflare Free App Stack","href":"/recipes/cloudflare-free-app-stack"},{"id":"archive/theme6-autonomous-incident-triage-agent","title":"Autonomous Incident Triage Agent for On-Call SREs","href":"/recipes/archive/theme6-autonomous-incident-triage-agent"},{"id":"cloudflare-ai-gateway-free-llm-fallback","title":"Cloudflare AI Gateway Free LLM Fallback Chain","href":"/recipes/cloudflare-ai-gateway-free-llm-fallback"},{"id":"vercel-agent-ai-gateway-central-hub","title":"Vercel Agent AI Gateway Central Hub","href":"/recipes/vercel-agent-ai-gateway-central-hub"},{"id":"archive/theme6-natural-language-iac-generator","title":"Natural-Language Infrastructure-as-Code Generator","href":"/recipes/archive/theme6-natural-language-iac-generator"},{"id":"archive/theme6-kubernetes-sre-helper-agent","title":"Kubernetes SRE Helper Agent for Multi-Cluster Toil","href":"/recipes/archive/theme6-kubernetes-sre-helper-agent"},{"id":"archive/theme0-langgraph-gmail-email-assistant","title":"Human-in-the-Loop Gmail Email Assistant","href":"/recipes/archive/theme0-langgraph-gmail-email-assistant"},{"id":"archive/theme5-langgraph-social-media-post-agent","title":"URL to Twitter & LinkedIn Posts with Human Review","href":"/recipes/archive/theme5-langgraph-social-media-post-agent"},{"id":"archive/theme7-langgraph-supervisor-research-desk","title":"LangGraph Supervisor Research Desk","href":"/recipes/archive/theme7-langgraph-supervisor-research-desk"},{"id":"archive/theme7-crewai-hierarchical-content-crew","title":"CrewAI Hierarchical Content Crew","href":"/recipes/archive/theme7-crewai-hierarchical-content-crew"},{"id":"archive/theme4-zendesk-webhook-triage-agent","title":"Zendesk Ticket Triage Agent: Classify, Search the KB, Route to Slack","href":"/recipes/archive/theme4-zendesk-webhook-triage-agent"},{"id":"archive/theme7-llamaindex-agentworkflow-report-swarm","title":"LlamaIndex AgentWorkflow Report Swarm","href":"/recipes/archive/theme7-llamaindex-agentworkflow-report-swarm"},{"id":"archive/theme4-llamaindex-document-agents-support","title":"Document Agents over Manuals and FAQs with LlamaIndex","href":"/recipes/archive/theme4-llamaindex-document-agents-support"},{"id":"archive/theme7-parallel-subagent-research-swarm","title":"Parallel Subagent Research Swarm","href":"/recipes/archive/theme7-parallel-subagent-research-swarm"},{"id":"archive/theme2-llamaindex-codebase-qa","title":"Chat With Your Codebase Using LlamaIndex RAG","href":"/recipes/archive/theme2-llamaindex-codebase-qa"},{"id":"archive/theme0-google-calendar-mcp-agent","title":"Natural-Language Google Calendar Agent (MCP)","href":"/recipes/archive/theme0-google-calendar-mcp-agent"},{"id":"archive/theme3-crewai-multi-agent-lead-scraper","title":"Multi-Agent Lead Scraper that Outputs a Structured Leads Table","href":"/recipes/archive/theme3-crewai-multi-agent-lead-scraper"},{"id":"archive/theme2-openhands-dependency-migration-agent","title":"Autonomous Dependency Migration Agent with OpenHands","href":"/recipes/archive/theme2-openhands-dependency-migration-agent"},{"id":"archive/theme1-firecrawl-deep-research-agent","title":"Deep Research Agent with Firecrawl and an Agent SDK","href":"/recipes/archive/theme1-firecrawl-deep-research-agent"},{"id":"archive/theme1-langgraph-exa-deep-research-pipeline","title":"Auditable Deep Research Pipeline with LangGraph and Exa","href":"/recipes/archive/theme1-langgraph-exa-deep-research-pipeline"},{"id":"archive/theme1-crewai-competitive-analysis-crew","title":"Competitive Analysis Crew that Outputs a Strategic Report","href":"/recipes/archive/theme1-crewai-competitive-analysis-crew"},{"id":"archive/theme1-searxng-private-web-research-agent","title":"Private Web Research Agent with Self-Hosted SearXNG","href":"/recipes/archive/theme1-searxng-private-web-research-agent"}]},{"id":"notification","name":"Notification","items":[{"id":"archive/theme6-kubernetes-sre-helper-agent","title":"Kubernetes SRE Helper Agent for Multi-Cluster Toil","href":"/recipes/archive/theme6-kubernetes-sre-helper-agent"},{"id":"archive/theme0-firecrawl-news-digest-email","title":"Daily Web Digest to Your Inbox (Firecrawl)","href":"/recipes/archive/theme0-firecrawl-news-digest-email"},{"id":"archive/theme0-triggerdev-per-user-morning-briefing","title":"Per-User Morning Briefing on Trigger.dev","href":"/recipes/archive/theme0-triggerdev-per-user-morning-briefing"},{"id":"archive/theme1-firecrawl-monitor-competitor-watch-agent","title":"Competitor Web-Change Watch Agent with Firecrawl Monitor","href":"/recipes/archive/theme1-firecrawl-monitor-competitor-watch-agent"}]},{"id":"knowledge-base","name":"Knowledge Base","items":[{"id":"archive/theme6-log-clustering-triage-agent","title":"Log Clustering and Triage Agent over MCP","href":"/recipes/archive/theme6-log-clustering-triage-agent"},{"id":"archive/theme6-scheduled-log-anomaly-sentinel","title":"Scheduled Log-Anomaly Sentinel with Vector Search","href":"/recipes/archive/theme6-scheduled-log-anomaly-sentinel"},{"id":"archive/theme4-langgraph-qdrant-hybrid-support-chatbot","title":"Hybrid-Search Customer Support Chatbot with LangGraph and Qdrant","href":"/recipes/archive/theme4-langgraph-qdrant-hybrid-support-chatbot"},{"id":"archive/theme4-firecrawl-docs-to-knowledge-base-agent","title":"Turn a Documentation Site into a Support Agent with Firecrawl and LangGraph","href":"/recipes/archive/theme4-firecrawl-docs-to-knowledge-base-agent"},{"id":"archive/theme5-qdrant-brand-aware-content-rag","title":"Brand-Aware Content Generator with RAG","href":"/recipes/archive/theme5-qdrant-brand-aware-content-rag"},{"id":"archive/theme4-llamaindex-document-agents-support","title":"Document Agents over Manuals and FAQs with LlamaIndex","href":"/recipes/archive/theme4-llamaindex-document-agents-support"},{"id":"archive/theme2-llamaindex-codebase-qa","title":"Chat With Your Codebase Using LlamaIndex RAG","href":"/recipes/archive/theme2-llamaindex-codebase-qa"},{"id":"archive/theme0-second-brain-notes-rag","title":"Second-Brain Notes Assistant (RAG over your vault)","href":"/recipes/archive/theme0-second-brain-notes-rag"},{"id":"archive/theme3-llamaparse-pdf-manuals-qdrant-hybrid-search","title":"Parse Complex PDF Manuals into a Qdrant Hybrid-Search Knowledge Base","href":"/recipes/archive/theme3-llamaparse-pdf-manuals-qdrant-hybrid-search"},{"id":"archive/theme3-firecrawl-crawl-docs-upstash-vector-rag","title":"Crawl a Documentation Site into an Upstash Vector RAG Index","href":"/recipes/archive/theme3-firecrawl-crawl-docs-upstash-vector-rag"},{"id":"archive/theme3-langchain-incremental-indexing-pgvector-sync","title":"Incrementally Sync a Source into a pgvector Index Without Re-Embedding","href":"/recipes/archive/theme3-langchain-incremental-indexing-pgvector-sync"},{"id":"archive/theme2-firecrawl-qdrant-migration-assistant","title":"Live-Docs Migration Assistant with Firecrawl and Qdrant","href":"/recipes/archive/theme2-firecrawl-qdrant-migration-assistant"},{"id":"archive/theme1-searxng-private-web-research-agent","title":"Private Web Research Agent with Self-Hosted SearXNG","href":"/recipes/archive/theme1-searxng-private-web-research-agent"}]}],"curatedHighlightsByCategory":{"llm-providers":{"collection":"infrastructure","category":"llm-providers","updatedAt":"2026-07-12","picks":[{"slug":"google-gemini","headline":"The default free frontier API","description":"The strongest standing free tier in the catalog: current Gemini frontier models, a real free quota that renews daily, and no credit card at any point. Tool calling, structured output, and a long context window are all on the free tier — the whole modern feature set, not a crippled demo. Note the two honest trade-offs: the API is Google-native REST rather than OpenAI-compatible, and free-tier prompts are used for model training with no opt-out. Rate limits vary by model and region, but for agents, prototypes, and side projects the free tier is genuinely workable as a daily driver. If you only configure one free LLM provider, make it this one.","whyChips":["No card","Frontier models","Daily quota"],"verified":{"verifiedAt":"2026-07-07","confidence":"medium","note":"Limits verified via freellm.net (official Google docs unreachable); ongoing free tier, no card confirmed; free-tier prompts train Google models (no opt-out); NOT OpenAI-compatible."},"offer":{"slug":"google-gemini","title":"Google AI Studio / Gemini API","provider":"Google AI Studio / Gemini API","planName":"freemium","logo":"/logos/gemini.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free tier exists; exact RPM/TPM/TPD varies by model, project, region, and live AI Studio limits.","keyMetrics":[{"key":"model","label":"Best model","value":"50.2 AA","display":"50.2 AA"},{"key":"requests_per_month","label":"Requests/month","value":"5K R/mo","display":"5K R/mo"}],"url":"https://ai.google.dev/gemini-api/docs"}},{"slug":"groq","headline":"Absurd speed, free to start","description":"LPU inference makes open models (Llama, Qwen, GPT-OSS) respond at hundreds of tokens per second — interactive agents feel instant. The free plan has honest per-model RPM/RPD/TPM limits visible in the console, no card required. When latency is the feature, start here.","whyChips":["Fastest inference","No card"],"verified":{"verifiedAt":"2026-07-07","confidence":"high","note":"All five S2.5 points live-verified (console rate-limit docs, pricing, ToS, model list); no card; commercial OK."},"offer":{"slug":"groq","title":"GroqCloud","provider":"GroqCloud","planName":"freemium","logo":"/logos/groq.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free plan limits are shown per organization/model in Groq Console; measured as RPM/RPD/TPM/TPD.","keyMetrics":[{"key":"model","label":"Best model","value":"23.8 AA","display":"23.8 AA"},{"key":"tokens_per_minute","label":"Tokens/min","value":"6K T/min","display":"6K T/min"},{"key":"requests_per_day","label":"Requests/day","value":"14.4K R/day","display":"14.4K R/day"}],"url":"https://console.groq.com/"}},{"slug":"openrouter","headline":"27 free models behind one key","description":"One OpenAI-compatible key routes to every provider's zero-price models — 27 free routes at last count, including strong recent open-weights. Free-model limits are request-based (about 20 RPM / 50 per day without a top-up), which is plenty for evaluation and light agent duty. The single-integration hedge against any one provider changing its terms.","whyChips":["One key, many models","OpenAI-compatible"],"verified":{"verifiedAt":"2026-07-07","confidence":"high","note":"Live-verified: 27+ :free routes, 20 RPM / 50 RPD (1,000 RPD with $10 credit), no card, commercial OK; credit balance must stay at $0 or above."},"offer":{"slug":"openrouter","title":"OpenRouter","provider":"OpenRouter","planName":"freemium","logo":"/logos/apis/openrouter.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"27 zero-price OpenRouter routes; no-top-up free-model limits are request-based (20 RPM, 50 requests/day). Buying credits raises the daily limit.","keyMetrics":[{"key":"model","label":"Best model","value":"37.8 AA","display":"37.8 AA"},{"key":"requests_per_day","label":"Requests/day","value":"50 R/day","display":"50 R/day"},{"key":"requests_per_minute","label":"Requests/min","value":"20 R/min","display":"20 R/min"}],"url":"https://openrouter.ai/"}}],"hero":{"slug":"google-gemini","headline":"The default free frontier API","description":"The strongest standing free tier in the catalog: current Gemini frontier models, a real free quota that renews daily, and no credit card at any point. Tool calling, structured output, and a long context window are all on the free tier — the whole modern feature set, not a crippled demo. Note the two honest trade-offs: the API is Google-native REST rather than OpenAI-compatible, and free-tier prompts are used for model training with no opt-out. Rate limits vary by model and region, but for agents, prototypes, and side projects the free tier is genuinely workable as a daily driver. If you only configure one free LLM provider, make it this one.","whyChips":["No card","Frontier models","Daily quota"],"verified":{"verifiedAt":"2026-07-07","confidence":"medium","note":"Limits verified via freellm.net (official Google docs unreachable); ongoing free tier, no card confirmed; free-tier prompts train Google models (no opt-out); NOT OpenAI-compatible."},"offer":{"slug":"google-gemini","title":"Google AI Studio / Gemini API","provider":"Google AI Studio / Gemini API","planName":"freemium","logo":"/logos/gemini.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free tier exists; exact RPM/TPM/TPD varies by model, project, region, and live AI Studio limits.","keyMetrics":[{"key":"model","label":"Best model","value":"50.2 AA","display":"50.2 AA"},{"key":"requests_per_month","label":"Requests/month","value":"5K R/mo","display":"5K R/mo"}],"url":"https://ai.google.dev/gemini-api/docs"}},"nominations":[{"slug":"mistral-ai","label":"Frontier free","description":"Frontier Mistral models (Medium 3.5, Large 3, Codestral) on a no-card Experiment tier that is ongoing, not a trial. Data training is on by default and the tier is framed for evaluation — flip the opt-out and it is the best EU-hosted free entry point.","verified":{"verifiedAt":"2026-07-10","confidence":"high","note":"Live-verified: ongoing no-card free tier, 10 models incl. frontier; exact limits login-gated; data-training default-on; commercial use requires paid Scale plan."},"icon":"medal","offer":{"slug":"mistral-ai","title":"Mistral AI","provider":"Mistral AI","planName":"freemium","logo":"/logos/mistral-ai.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free 'Experiment' plan, no credit card; public Mistral docs state limited free-mode rate limits and a token/month cap, but do not publish a numeric monthly token allowance. Prompts may be used to improve models.","keyMetrics":[{"key":"model","label":"Best model","value":"12.5 AA","display":"12.5 AA"}],"url":"https://console.mistral.ai/api-keys"}},{"slug":"ollama-cloud","label":"Privacy first","description":"A genuine $0 tier metered by GPU time rather than tokens, with frontier open models and a no-logging, no-training requirement on every partner. The only shortlist entry built around data privacy.","verified":{"verifiedAt":"2026-07-09","confidence":"high","note":"Live pricing page: $0 plan, GPU-time metering, 5h/7d limit resets, privacy/no-logging language; commercial allowed."},"icon":"medal","offer":{"slug":"ollama-cloud","title":"Ollama Cloud","provider":"Ollama Cloud","planName":"freemium","logo":"/logos/ollama-cloud.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free $0 plan includes access to cloud models with reported 'light usage' limits: session limits reset every 5 hours and weekly limits every 7 days; usage measured by GPU time, not tokens. Pro ($20/mo) and Max ($100/mo) raise limits 50x/250x.","keyMetrics":[{"key":"model","label":"Best model","value":"33.8 AA","display":"33.8 AA"}],"url":"https://ollama.com/cloud"}},{"slug":"fireworks-ai","label":"Model breadth","description":"34 models — DeepSeek V4, Llama 4, Qwen 3.6, Kimi K2, GLM-5.2 — behind one OpenAI-compatible key, with a $1 no-card signup credit that covers real evaluation. Production continues seamlessly as pay-as-you-go.","verified":{"verifiedAt":"2026-07-09","confidence":"high","note":"Live pricing page: $1 one-time signup credit, no card, instant key, OpenAI-compatible; commercial allowed; finite credit, not ongoing-free."},"icon":"medal","offer":{"slug":"fireworks-ai","title":"Fireworks AI","provider":"Fireworks AI","planName":"finite_credit","logo":"/logos/apis/fireworks-ai.png","pricingModel":"finite_credit","accessBucket":"conditional","freeLimitText":"$1 in free credits on signup for serverless text and vision models; pay-as-you-go thereafter. Finite, one-time.","keyMetrics":[{"key":"model","label":"Best model","value":"51.1 AA","display":"51.1 AA"},{"key":"requests_per_minute","label":"Requests/min","value":"10 R/min","display":"10 R/min"},{"key":"credit_amount_usd","label":"Credit","value":"1 USD","display":"1 USD"}],"url":"https://fireworks.ai/"}}],"paidPicks":[{"slug":"kimi-code-moderato","description":"A focused coding subscription for Kimi's long-context code models with OpenAI and Anthropic-compatible endpoints.","priceFact":"$19/mo","offer":{"slug":"kimi-code-moderato","title":"Kimi Code Moderato","provider":"Kimi Code Moderato","planName":"paid","logo":"/logos/kimi-code-moderato.png","pricingModel":"paid","accessBucket":"paid","freeLimitText":"Paid — $19/mo ($15/mo billed annually). Kimi K2.7 Code + K2.7 HighSpeed (~180 tok/s) + K2.6; ~300-1,200 requests/5h with weekly quota, up to 30 concurrent; OpenAI- AND Anthropic-compatible endpoints (Claude Code, OpenCode, OpenClaw, custom agents)","keyMetrics":[{"key":"model","label":"Best model","value":"42 AA","display":"42 AA"},{"key":"requests_per_month","label":"Requests/month","value":"172.8K R/mo","display":"172.8K R/mo"},{"key":"paid_value_ratio","label":"Value ratio","value":"0.05 value/$","display":"0.05 value/$"}],"url":"https://www.kimi.com/code"}},{"slug":"command-code-provider-api","description":"A direct provider API with included usage, discounted open models, and list-price access to premium frontier models.","priceFact":"$15/mo + PAYG","offer":{"slug":"command-code-provider-api","title":"Command Code Provider API","provider":"Command Code Provider API","planName":"paid","logo":"/logos/command-code-provider-api.png","pricingModel":"paid","accessBucket":"paid","freeLimitText":"Paid — $15/mo ($15 usage included, then PAYG at model rates, no markup). Raw OpenAI- + Anthropic-compatible API ('no lock-in — any client or agent') with GLM-5.2 + GLM-5.2 Fast (1M ctx), Kimi K2.7 Code + HighSpeed, DeepSeek V4 Pro (75% off), MiMo V2.5 (98% off), plus Claude/GPT/Gemini at list; $15 usage included, then pay-as-you-go with permanent discounts","keyMetrics":[{"key":"model","label":"Best model","value":"51.1 AA","display":"51.1 AA"}],"url":"https://commandcode.ai/pricing"}},{"slug":"deepseek-platform-api","description":"Low-cost metered DeepSeek API access for chat and reasoning workloads when pay-as-you-go pricing is the right fit.","priceFact":"$0.28/M out","offer":{"slug":"deepseek-platform-api","title":"DeepSeek Platform API","provider":"DeepSeek Platform API","planName":"paid","logo":"/logos/deepseek-platform-api.png","pricingModel":"paid","accessBucket":"paid","freeLimitText":"Paid — Pay-as-you-go, ~$0.28/M output (chat); R1 $0.55/$2.19. deepseek-chat (V3.2): $0.028 cache / $0.28 input miss / $0.42 output per 1M; deepseek-reasoner (R1): $0.14/$0.55/$2.19, cheapest top-tier reasoning","keyMetrics":[{"key":"model","label":"Best model","value":"24.7 AA","display":"24.7 AA"}],"url":"https://api-docs.deepseek.com/quick_start/pricing"}}],"specialSlot":{"slug":"nvidia","description":"NVIDIA NIM is a conditional developer-access offer: useful when its model catalog and rate limits match the prototype.","priceFact":"40 RPM","verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Live-verified: ongoing rate-limited trial (~40 RPM), no card, instant NGC key, 145+ models; commercial use NOT allowed on the free endpoint (NVIDIA API Trial ToS)."},"label":"40 RPM developer access","offer":{"slug":"nvidia","title":"NVIDIA NIM","provider":"NVIDIA NIM","planName":"freemium","logo":"/logos/nvidia.svg","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free developer access: sign up and call many models through one endpoint; free use is governed by a 40 requests-per-minute rate limit (NVIDIA retired the old signup-credit system). Verify current rate limits and model list on the site.","keyMetrics":[{"key":"model","label":"Best model","value":"44.4 AA","display":"44.4 AA"},{"key":"requests_per_minute","label":"Requests/min","value":"40 R/min","display":"40 R/min"}],"url":"https://build.nvidia.com/"}}},"embeddings":{"collection":"infrastructure","category":"embeddings","updatedAt":"2026-07-10","intro":"Hosted vector and embedding services with free starter capacity for retrieval-augmented agent workflows.","picks":[{"slug":"pinecone","headline":"Mainstream managed vector search","description":"Pinecone is the hero embeddings pick because it is the most recognized managed vector database with a clear and generous free Starter plan. The free tier includes up to 2 GB storage, 2 million write units per month, 1 million read units per month, and up to 5 serverless indexes on AWS us-east-1, plus 5 million embedding tokens per month of inference and 1 GB of Assistant storage. That quota handles real retrieval prototypes and small production workloads before the paid Builder plan at $20 per month is needed.","whyChips":["2 GB storage","Serverless indexes","Inference included"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"pinecone","title":"Pinecone","provider":"Pinecone","planName":"freemium","logo":"/logos/pinecone-builder.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"The Starter plan is free with no credit card: up to 2 GB storage, 2M write units/mo, 1M read units/mo, and up to 5 serverless indexes on AWS us-east-1, plus 5M embedding tokens/mo of Inference and 1 GB of Assistant storage.","keyMetrics":[{"key":"free_tokens_per_month","label":"Monthly tokens","value":"5M T/mo","display":"5M T/mo"},{"key":"storage_gb","label":"Storage","value":"2 GB","display":"2 GB"}],"url":"https://www.pinecone.io/"}},{"slug":"weaviate-cloud-serverless","headline":"Free sandbox with lowest storage cost","description":"Weaviate Cloud Serverless is the side pick for teams that want a managed Weaviate instance without running the stack themselves. The free tier includes 100,000 objects and 1 GB of storage, which is enough to evaluate Weaviate's hybrid search and vector capabilities. Beyond the free tier, the per-dimension pricing at $0.095 per million dimensions is among the cheapest published rates for vector storage.","whyChips":["100k objects free","Hybrid search","$0.095/M dims"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"weaviate-cloud-serverless","title":"Weaviate Cloud","provider":"Weaviate Cloud","planName":"freemium","logo":"/logos/weaviate-cloud-serverless.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free tier (100K objects, 1 GB) to start; Serverless then ~$25/mo minimum, then $0.095 per 1M vector dimensions/month (storage) — among the cheapest per-dimension vector rates.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"1 GB","display":"1 GB"}],"url":"https://weaviate.io/pricing"}},{"slug":"zilliz-cloud","headline":"Managed Milvus with largest free quota","description":"Zilliz Cloud is the side pick for teams that want the largest free quota among the managed Milvus options. The free cluster includes 5 GB storage (approximately 1 million 768-dimensional vectors), 2.5 million vCUs per month, and up to 5 collections with no credit card required. As the creators of Milvus, the migration story and ecosystem support are clearer than for other managed vector stores.","whyChips":["5 GB storage","Milvus ecosystem","No card"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"zilliz-cloud","title":"Zilliz Cloud","provider":"Zilliz Cloud","planName":"freemium","logo":"/logos/zilliz-cloud-serverless.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"One Free cluster per organization with 5 GB storage (~1M 768-dim vectors), 2.5M vCUs/month, and up to 5 collections, no credit card required.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"5 GB","display":"5 GB"}],"url":"https://zilliz.com"}}],"hero":{"slug":"pinecone","headline":"Mainstream managed vector search","description":"Pinecone is the hero embeddings pick because it is the most recognized managed vector database with a clear and generous free Starter plan. The free tier includes up to 2 GB storage, 2 million write units per month, 1 million read units per month, and up to 5 serverless indexes on AWS us-east-1, plus 5 million embedding tokens per month of inference and 1 GB of Assistant storage. That quota handles real retrieval prototypes and small production workloads before the paid Builder plan at $20 per month is needed.","whyChips":["2 GB storage","Serverless indexes","Inference included"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"pinecone","title":"Pinecone","provider":"Pinecone","planName":"freemium","logo":"/logos/pinecone-builder.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"The Starter plan is free with no credit card: up to 2 GB storage, 2M write units/mo, 1M read units/mo, and up to 5 serverless indexes on AWS us-east-1, plus 5M embedding tokens/mo of Inference and 1 GB of Assistant storage.","keyMetrics":[{"key":"free_tokens_per_month","label":"Monthly tokens","value":"5M T/mo","display":"5M T/mo"},{"key":"storage_gb","label":"Storage","value":"2 GB","display":"2 GB"}],"url":"https://www.pinecone.io/"}},"nominations":[{"slug":"qdrant-cloud-free","label":"Small cluster","description":"Qdrant Cloud Free Tier provides a free-forever single-node cluster with 0.5 vCPU, 1 GB RAM, and 4 GB disk, plus free cloud inference on selected models. It is the nomination for developers who want a straightforward managed vector database with a generous free tier and clear upgrade path to dedicated infrastructure.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"qdrant-cloud-free","title":"Qdrant Cloud Free Tier","provider":"Qdrant Cloud Free Tier","planName":"free","logo":"/logos/qdrant-cloud-free.svg","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free forever: single-node cluster with 0.5 vCPU, 1GB RAM, 4GB disk, plus free cloud inference with selected models.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"4 GB","display":"4 GB"}],"url":"https://qdrant.tech/pricing/"}},{"slug":"upstash-vector","label":"Serverless usage","description":"Upstash Vector is the nomination for serverless applications that want vector search with per-request billing and daily operation limits. The free tier includes 10,000 daily query and update operations with up to 1 GB of data, making it suitable for lightweight retrieval workloads where cost predictability matters.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"upstash-vector","title":"Upstash | Vector","provider":"Upstash","planName":"freemium","logo":"/logos/upstash.svg","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free tier: 10K daily query/update operations, 1 GB max data/metadata size, 1,536 max dimensions, and 100 namespaces.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"1 GB","display":"1 GB"}],"url":"https://upstash.com/docs/vector"}}],"paidPicks":[]},"databases":{"collection":"infrastructure","category":"databases","updatedAt":"2026-07-10","intro":"Managed databases with durable free or conditional capacity.","picks":[{"slug":"supabase","headline":"The full backend Postgres default","description":"Supabase is the hero database pick because it pairs a managed Postgres instance with auth, file storage, real-time subscriptions, and edge functions in one project. The free tier gives a workable project for prototypes and small products without a credit card. It is the lowest-friction path from zero to a complete hosted backend.","whyChips":["No card","Full backend","Auth + storage"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"supabase","title":"Supabase","provider":"Supabase","planName":"freemium","logo":"/logos/supabase-pro.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free project tier; storage/database quotas depend on current plan.","keyMetrics":[{"key":"bandwidth_gb","label":"Egress","value":"5 GB/mo","display":"5 GB/mo"},{"key":"max_databases","label":"Databases","value":"2 DBs","display":"2 DBs"}],"url":"https://supabase.com/"}},{"slug":"neon","headline":"Non-branchable serverless Postgres","description":"Neon is a strong side pick because it separates compute and storage, giving you serverless Postgres that scales to zero without cold-start latency on every connection. Branching databases for staging and preview environments is a distinctive feature that fits agent development workflows where each agent run needs an isolated data view.","whyChips":["Scales to zero","Branching","Serverless Postgres"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"neon","title":"Neon","provider":"Neon","planName":"free_forever","logo":"/logos/neon-launch.png","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Free tier available; compute/storage limits depend on plan.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"0.5 GB","display":"0.5 GB"},{"key":"bandwidth_gb","label":"Egress","value":"5 GB/mo","display":"5 GB/mo"},{"key":"compute_hours","label":"Compute","value":"100 hrs/mo","display":"100 hrs/mo"}],"url":"https://neon.com/"}},{"slug":"turso","headline":"SQLite close to the edge","description":"Turso is the side pick for applications that benefit from embedded SQLite semantics replicated to the edge. The free starter tier covers small production experiments and agent state use cases. The libSQL compatibility also provides a familiar interface for Postgres migrants who want multi-tenant isolation without the overhead of a full Postgres cluster.","whyChips":["Edge SQLite","libSQL","Free starter"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"turso","title":"Turso","provider":"Turso","planName":"freemium","logo":"/logos/turso-developer.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free starter tier available; rows/storage/groups vary by plan.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"5 GB","display":"5 GB"},{"key":"row_writes_m","label":"Writes","value":"10 M rows/mo","display":"10 M rows/mo"},{"key":"bandwidth_gb","label":"Egress","value":"3 GB/mo","display":"3 GB/mo"}],"url":"https://turso.tech/"}}],"hero":{"slug":"supabase","headline":"The full backend Postgres default","description":"Supabase is the hero database pick because it pairs a managed Postgres instance with auth, file storage, real-time subscriptions, and edge functions in one project. The free tier gives a workable project for prototypes and small products without a credit card. It is the lowest-friction path from zero to a complete hosted backend.","whyChips":["No card","Full backend","Auth + storage"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"supabase","title":"Supabase","provider":"Supabase","planName":"freemium","logo":"/logos/supabase-pro.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free project tier; storage/database quotas depend on current plan.","keyMetrics":[{"key":"bandwidth_gb","label":"Egress","value":"5 GB/mo","display":"5 GB/mo"},{"key":"max_databases","label":"Databases","value":"2 DBs","display":"2 DBs"}],"url":"https://supabase.com/"}},"nominations":[{"slug":"cockroachdb-basic","label":"Scale-to-zero","description":"CockroachDB Basic includes 50 million RUs and 10 GiB storage free per month on the Basic plan, scales to zero between uses, and requires no credit card. It is a practical way to get a globally resilient, Postgres-compatible SQL database without managing infrastructure or committing to a paid plan.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"cockroachdb-basic","title":"CockroachDB Cloud","provider":"CockroachDB Cloud","planName":"freemium","logo":"/logos/cockroachdb-basic.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"50 million Request Units (RUs) and 10 GiB storage free per month on the Basic plan; scales to zero; no credit card required for Basic. ($400 in trial credits also available separately.)","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"10 GB","display":"10 GB"},{"key":"max_databases","label":"Databases","value":"1 DBs","display":"1 DBs"}],"url":"https://www.cockroachlabs.com/pricing/"}},{"slug":"prisma-postgres","label":"Postgres in one click","description":"Prisma Postgres is directly integrated with the Prisma ORM, making it a natural choice for teams already using Prisma for schema management. The free tier includes 100,000 operations per month, 500 MB storage, and 50 databases with no card required, which is enough to evaluate a small project before committing to a paid plan.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"prisma-postgres","title":"Prisma Postgres","provider":"Prisma Postgres","planName":"freemium","logo":"/logos/prisma-postgres-starter.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free: 100,000 operations/month included, 500 MB storage, 50 databases, no credit card required.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"0.5 GB","display":"0.5 GB"},{"key":"bandwidth_gb","label":"Egress","value":"∞","display":"∞"},{"key":"max_databases","label":"Databases","value":"50 DBs","display":"50 DBs"}],"url":"https://www.prisma.io/pricing"}}],"paidPicks":[]},"app-platforms":{"collection":"infrastructure","category":"app-platforms","updatedAt":"2026-07-10","intro":"Full-stack and backend hosting with credible free deploy paths.","picks":[{"slug":"convex-free","headline":"The reactive backend-in-a-box","description":"Convex is the hero app-platform pick because it combines a reactive Postgres-like database, serverless functions, scheduled jobs, auth, file storage, and vector search into one coherent platform. The free and Starter tiers cover up to 40 deployments and 1 million function calls per month, which is enough to build and ship a real agent app without stitching together separate services. The reactive programming model reduces the boilerplate of keeping agent state synchronized across clients.","whyChips":["Reactive database","Functions + cron","1M calls/mo"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"convex-free","title":"Convex","provider":"Convex","planName":"free","logo":"/logos/convex-free.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free & Starter: reactive database, file/text/vector search, crons, auth, Node.js actions, up to 40 deployments (1M function calls/month).","keyMetrics":[{"key":"bandwidth_gb_mo","label":"Bandwidth","value":"1 GB/mo","display":"1 GB/mo"}],"url":"https://convex.dev/pricing"}},{"slug":"koyeb-free","headline":"Simple container platform plus Postgres","description":"Koyeb is the side pick for teams that want a simple container platform with a managed Postgres option and a path to serverless GPUs without Kubernetes complexity. The free tier gives one web service with 512 MB RAM, 0.1 vCPU, 2 GB SSD, plus a free PostgreSQL database with 1 GB storage and 100 GB monthly outbound bandwidth. Per-second billing means you only pay for what you use once production traffic starts.","whyChips":["Containers + Postgres","GPU path","Free tier"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"koyeb-free","title":"Koyeb","provider":"Koyeb","planName":"free","logo":"/logos/koyeb.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"One free web Service (512 MB RAM, 0.1 vCPU, 2 GB SSD) in Frankfurt, Washington D.C., or Singapore, plus one free PostgreSQL database (1 GB storage, 5 hours active time), and 100 GB free outbound bandwidth/month. Requires a credit card (a $29 pre-authorization hold is placed during verification).","keyMetrics":[{"key":"bandwidth_gb_mo","label":"Bandwidth","value":"100 GB/mo","display":"100 GB/mo"},{"key":"build_minutes_mo","label":"Build minutes","value":"∞","display":"∞"},{"key":"custom_domains","label":"Custom domains","value":"5 domains","display":"5 domains"}],"url":"https://www.koyeb.com/docs/faqs/pricing"}},{"slug":"huggingface-spaces","headline":"Best place to publish an AI demo","description":"Hugging Face Spaces is the side pick for agent apps that need to ship an AI-facing demo to collaborators or users quickly. Free CPU Spaces exist and the community discovery on the platform gives prototypes visibility without any hosting setup. The trade-off is that free CPU Spaces sleep and restart, so they are not reliable for always-on bots or real-time agents — but for shareable demos, evaluations, and prototypes, the path from zero to a live demo is the shortest available.","whyChips":["AI demos","Community discovery","Free CPU"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"huggingface-spaces","title":"Hugging Face Spaces","provider":"Hugging Face Spaces","planName":"freemium","logo":"/logos/huggingface.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free CPU Spaces exist; sleeping/restarts make realtime bots unreliable.","keyMetrics":[{"key":"bandwidth_gb_mo","label":"Bandwidth","value":"∞","display":"∞"},{"key":"build_minutes_mo","label":"Build minutes","value":"∞","display":"∞"}],"url":"https://huggingface.co/spaces"}}],"hero":{"slug":"convex-free","headline":"The reactive backend-in-a-box","description":"Convex is the hero app-platform pick because it combines a reactive Postgres-like database, serverless functions, scheduled jobs, auth, file storage, and vector search into one coherent platform. The free and Starter tiers cover up to 40 deployments and 1 million function calls per month, which is enough to build and ship a real agent app without stitching together separate services. The reactive programming model reduces the boilerplate of keeping agent state synchronized across clients.","whyChips":["Reactive database","Functions + cron","1M calls/mo"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"convex-free","title":"Convex","provider":"Convex","planName":"free","logo":"/logos/convex-free.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free & Starter: reactive database, file/text/vector search, crons, auth, Node.js actions, up to 40 deployments (1M function calls/month).","keyMetrics":[{"key":"bandwidth_gb_mo","label":"Bandwidth","value":"1 GB/mo","display":"1 GB/mo"}],"url":"https://convex.dev/pricing"}},"nominations":[{"slug":"leapcell","label":"Always-on","description":"Leapcell's Hobby plan covers up to 20 projects with 3 vCPU and 4 GB RAM each, 100,000 serverless invocations, 3 GB-hours compute, 1 free PostgreSQL database, 100K Redis commands per month, and 2 GB egress per month — all at no cost with no credit card required. The offering is distinctive because it includes a real always-on Postgres alongside serverless functions, which is rare at the free tier.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"leapcell","title":"Leapcell","provider":"Leapcell","planName":"freemium","logo":"/logos/leapcell.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Hobby plan ($0): up to 20 projects (3 vCPU / 4 GB RAM each), 100,000 serverless invocations and 3 GB-hours compute per month, 1 free PostgreSQL database, 100K Redis commands/mo, 2 GB egress/mo, 60 build min/mo, plus custom domains with free TLS and global CDN.","keyMetrics":[{"key":"bandwidth_gb_mo","label":"Bandwidth","value":"2 GB/mo","display":"2 GB/mo"},{"key":"build_minutes_mo","label":"Build minutes","value":"60 min/mo","display":"60 min/mo"}],"url":"https://leapcell.io"}},{"slug":"nhost-free","label":"Full backend","description":"Nhost is the nomination for teams that want a GraphQL-first managed backend with Postgres, Hasura, auth, storage, and serverless functions already wired together. The free Starter plan includes 1 GB database, 1 GB storage, 5 GB egress, and unlimited users — enough for a small production experiment or a weekend project that does not want to configure each service individually.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"nhost-free","title":"Nhost","provider":"Nhost","planName":"free","logo":"/logos/nhost-free.svg","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Starter $0/month: 1 GB database, 1 GB storage, 5 GB egress, functions, realtime APIs, unlimited users, OAuth, 1 project (paused after 1 week idle).","keyMetrics":[{"key":"bandwidth_gb_mo","label":"Bandwidth","value":"5 GB/mo","display":"5 GB/mo"},{"key":"function_invocations_mo","label":"Function calls","value":"∞","display":"∞"}],"url":"https://nhost.io/pricing"}}],"paidPicks":[]},"agent-deployment":{"collection":"infrastructure","category":"agent-deployment","updatedAt":"2026-07-12","intro":"Hosted agent runtimes with useful free or starter capacity.","picks":[{"slug":"cloudflare-agents","headline":"Best free stateful agent runtime","description":"Cloudflare Agents is the strongest default pick for a hosted agent runtime because it combines Workers, Durable Objects, and a large no-card free allocation. The free plan gives enough daily request and stateful compute capacity for prototypes, internal tools, and small public agents. It is also close to the edge platform many teams already use, so the path from experiment to production is short.","whyChips":["No card","Stateful runtime","Edge deploys"],"verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Workers Free 100k req/day + DO 100k req/day / 13,000 GB-s/day confirmed live on the Durable Objects pricing page; no card."},"offer":{"slug":"cloudflare-agents","title":"Cloudflare Agents","provider":"Cloudflare Agents","planName":"free","logo":"/logos/cloudflare.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Workers Free plan: 100,000 requests/day, 10 ms CPU/invocation, unlimited bandwidth. Durable Objects (which back each agent's state) are included free: 100,000 requests/day and 13,000 GB-s/day of compute. No credit card required. Workers AI inference has a separate free allocation.","keyMetrics":[{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"4 signals","display":"4 signals"},{"key":"agent_production_controls","label":"Production controls","value":"1 signals","display":"1 signals"}],"url":"https://agents.cloudflare.com"}},{"slug":"lyzr-agent-studio","headline":"Generous no-code agent studio","description":"Lyzr is a useful pick when the goal is to build and test multiple hosted agents without starting from infrastructure code. The community tier lists meaningful monthly credits, agent slots, and knowledge-base capacity. It is a good editorial counterpoint to code-first runtimes.","whyChips":["Hosted studio","Monthly credits"],"verified":{"verifiedAt":"2026-07-12","confidence":"medium","note":"Community Plan ($0, 10 agents, 500 credits/mo) confirmed in docs.lyzr.ai/introduction/plans, but the main pricing page now leads with per-run pricing only."},"offer":{"slug":"lyzr-agent-studio","title":"Lyzr Agent Studio","provider":"Lyzr Agent Studio","planName":"free","logo":"/logos/lyzr-agent-studio.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Ongoing free tier on Lyzr Agent Studio (hosted Lyzr Cloud). Two figures appear in Lyzr's own docs: the Plans page lists a \"Community Plan\" at $0/month with 500 credits/month (1 credit = $1), build up to 10 agents, 5 knowledge bases, 100 MB RAG storage, 5 tools, 1 builder license, 7-day logs; the Cre","keyMetrics":[{"key":"agent_execution_quota","label":"Execution quota","value":"1 runs/month","display":"1 runs/month"},{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"1 signals","display":"1 signals"},{"key":"agent_production_controls","label":"Production controls","value":"1 signals","display":"1 signals"}],"url":"https://www.lyzr.ai"}},{"slug":"openai-agentkit-agent-builder-chatkit-agents-sdk","headline":"Best builder experience","description":"OpenAI AgentKit is notable for Agent Builder, ChatKit, and the Agents SDK living in one workflow. The builder surface is free to use, while runtime cost follows model usage. Pick it when design, handoff, and SDK alignment matter more than raw free compute.","whyChips":["Visual builder","SDK aligned"],"verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Agent Builder/ChatKit/Agents SDK included with standard API model pricing (openai.com AgentKit announcement); runtime = paid tokens."},"offer":{"slug":"openai-agentkit-agent-builder-chatkit-agents-sdk","title":"OpenAI AgentKit","provider":"OpenAI AgentKit","planName":"limited","logo":"/logos/openai-agentkit-agent-builder-chatkit-agents-sdk.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Agent Builder canvas is free to use; you pay only for model/token usage when agents run","keyMetrics":[{"key":"agent_execution_quota","label":"Execution quota","value":"1 runs/month","display":"1 runs/month"},{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"6 signals","display":"6 signals"},{"key":"agent_production_controls","label":"Production controls","value":"1 signals","display":"1 signals"}],"url":"https://openai.com/index/introducing-agentkit/"}}],"hero":{"slug":"cloudflare-agents","headline":"Best free stateful agent runtime","description":"Cloudflare Agents is the strongest default pick for a hosted agent runtime because it combines Workers, Durable Objects, and a large no-card free allocation. The free plan gives enough daily request and stateful compute capacity for prototypes, internal tools, and small public agents. It is also close to the edge platform many teams already use, so the path from experiment to production is short.","whyChips":["No card","Stateful runtime","Edge deploys"],"verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Workers Free 100k req/day + DO 100k req/day / 13,000 GB-s/day confirmed live on the Durable Objects pricing page; no card."},"offer":{"slug":"cloudflare-agents","title":"Cloudflare Agents","provider":"Cloudflare Agents","planName":"free","logo":"/logos/cloudflare.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Workers Free plan: 100,000 requests/day, 10 ms CPU/invocation, unlimited bandwidth. Durable Objects (which back each agent's state) are included free: 100,000 requests/day and 13,000 GB-s/day of compute. No credit card required. Workers AI inference has a separate free allocation.","keyMetrics":[{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"4 signals","display":"4 signals"},{"key":"agent_production_controls","label":"Production controls","value":"1 signals","display":"1 signals"}],"url":"https://agents.cloudflare.com"}},"nominations":[{"slug":"modal","label":"Compute credits","description":"Starter credits and strong serverless primitives make Modal a practical choice for agent jobs that need containers or GPUs.","verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Starter $0/mo with $30/month recurring free credits, 100 containers + 10 GPU concurrency confirmed live on modal.com/pricing."},"icon":"medal","offer":{"slug":"modal","title":"Modal","provider":"Modal","planName":"paid","logo":"/logos/modal.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Paid — $0/mo (Starter) + usage; CPU $0.0000131/core-sec. Starter free with $30/mo free compute credits, 100 containers + 10 GPU concurrency, 5 crons, 8 webhooks. Sandbox CPU $0.00003942/core-sec, memory $0.00000672/GiB-sec, volumes $0.09/GiB-mo (1 TiB free). Team $250/mo + comput","keyMetrics":[{"key":"agent_execution_quota","label":"Execution quota","value":"1 runs/month","display":"1 runs/month"},{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"1 signals","display":"1 signals"},{"key":"agent_production_controls","label":"Production controls","value":"2 signals","display":"2 signals"}],"url":"https://modal.com/pricing"}},{"slug":"beam-cloud","label":"GPU starter","description":"Beam Cloud stands out for a developer tier with recurring credits and simple container deployment for AI workloads.","verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Developer tier $0/mo with $30 monthly credits and 5 GPU container concurrency confirmed live on beam.cloud/pricing."},"icon":"medal","offer":{"slug":"beam-cloud","title":"Beam Cloud","provider":"Beam Cloud","planName":"paid","logo":"/logos/beam-cloud.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Paid — $0/mo (Developer) + usage; CPU $0.0000528/core-sec. Developer free with $30/mo credits, unlimited apps, 5 GPU + 30 CPU containers. CPU $0.0000528/core-sec, RAM $0.0000056/GB-sec, RTX 4090 $0.000192/sec; on-demand H100 $1.74/hr, A100 80GB $1.30/hr; storage free; cold-start ","keyMetrics":[{"key":"agent_execution_quota","label":"Execution quota","value":"1 runs/month","display":"1 runs/month"},{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"1 signals","display":"1 signals"},{"key":"agent_production_controls","label":"Production controls","value":"1 signals","display":"1 signals"}],"url":"https://www.beam.cloud/pricing"}},{"slug":"arcade-dev","label":"Tool execution","description":"Arcade.dev is worth tracking for hosted tool execution, user authorization, and MCP-oriented agent integrations.","verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Hobby FREE: 1,000 standard + 50 pro tool executions, 100 auth challenges, 1 hosted MCP server 24/7 confirmed live on arcade.dev/pricing."},"icon":"medal","offer":{"slug":"arcade-dev","title":"Arcade.dev","provider":"Arcade.dev","planName":"free","logo":"/logos/arcade-dev.png","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Hobby plan (FREE, $0 platform fee, ongoing): 1,000 standard tool executions, 50 pro tool executions, 100 user authorization challenges, and 1 Arcade-hosted MCP server running 24/7 included. Shared tenant. Community support via GitHub Discussions. Rate limit 1,000/min.","keyMetrics":[{"key":"agent_execution_quota","label":"Execution quota","value":"1 runs/month","display":"1 runs/month"},{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"4 signals","display":"4 signals"},{"key":"agent_production_controls","label":"Production controls","value":"1 signals","display":"1 signals"}],"url":"https://www.arcade.dev/pricing"}}],"paidPicks":[]},"serverless-functions":{"collection":"infrastructure","category":"serverless-functions","updatedAt":"2026-07-10","intro":"Serverless execution with practical free developer capacity at the edge and beyond.","picks":[{"slug":"cloudflare-workers","headline":"The edge function default","description":"Cloudflare Workers is the hero serverless pick because it runs JavaScript and TypeScript at the edge with a genuinely useful free tier and a pricing model that stays simple as usage grows. The runtime is fast, the cold-start is near-zero, and Workers are tightly integrated with Cloudflare's KV, D1, R2, and Queues — which means agent logic can stay within one provider's ecosystem without sacrificing performance. The 100,000 requests per day free allowance covers meaningful prototype and hobby workloads.","whyChips":["100k req/day","Edge runtime","KV + D1 + R2"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"cloudflare-workers","title":"Cloudflare Workers","provider":"Cloudflare Workers","planName":"freemium","logo":"/logos/cloudflare-workers-paid.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free tier historically centered around daily request quotas; verify current limits before launch.","keyMetrics":[{"key":"max_cpu_per_invocation","label":"CPU/call","value":"10 ms","display":"10 ms"},{"key":"bandwidth_per_month","label":"Bandwidth","value":"∞","display":"∞"}],"url":"https://developers.cloudflare.com/workers/"}},{"slug":"val-town","headline":"Fastest script-to-endpoint path","description":"Val Town earns its pick as the fastest path from a script to a hosted endpoint. The free plan gives 100,000 runs per day at 15-minute intervals, unlimited public vals, and MCP integration — enough to build cron jobs, webhooks, and small AI-connected endpoints without any deployment pipeline. The wall-clock limit and public-only constraint on the free plan are honest trade-offs that match the simplicity of the use case.","whyChips":["100k runs/day","No deploy","MCP integration"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"val-town","title":"Val Town","provider":"Val Town","planName":"freemium","logo":"/logos/val-town.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free plan: 100,000 runs/day, unlimited public vals, 15-minute cron intervals, 1-minute wall-clock time per run, 3-day log retention, MCP integration. (Free tier allows 0 private vals.)","keyMetrics":[{"key":"compute_gb_seconds","label":"Compute","value":"∞","display":"∞"},{"key":"max_cpu_per_invocation","label":"CPU/call","value":"60K ms","display":"60K ms"},{"key":"bandwidth_per_month","label":"Bandwidth","value":"∞","display":"∞"}],"url":"https://www.val.town/pricing"}},{"slug":"deno-deploy","headline":"Modern TypeScript at the edge","description":"Deno Deploy is the side pick for teams that want a first-class TypeScript runtime with built-in KV and volume storage at the edge. The free plan includes 1 million inbound requests per month, 20 GB egress, 1 GiB volume storage, and Deno KV with 450k read and 300k write units per month — a generous allowance for a modern edge-native stack that avoids Node.js compatibility overhead.","whyChips":["TypeScript edge","KV included","1M req + 20GB egress"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"deno-deploy","title":"Deno Deploy","provider":"Deno Deploy","planName":"freemium","logo":"/logos/deno-deploy-pro.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free plan: 1M inbound requests/month, 20 GB egress, 1 GiB volume storage, 1 GiB Deno KV storage, 450k KV read units and 300k KV write units per month, 50 custom domains.","keyMetrics":[{"key":"bandwidth_per_month","label":"Bandwidth","value":"20 GB/mo","display":"20 GB/mo"}],"url":"https://deno.com/deploy/pricing"}}],"hero":{"slug":"cloudflare-workers","headline":"The edge function default","description":"Cloudflare Workers is the hero serverless pick because it runs JavaScript and TypeScript at the edge with a genuinely useful free tier and a pricing model that stays simple as usage grows. The runtime is fast, the cold-start is near-zero, and Workers are tightly integrated with Cloudflare's KV, D1, R2, and Queues — which means agent logic can stay within one provider's ecosystem without sacrificing performance. The 100,000 requests per day free allowance covers meaningful prototype and hobby workloads.","whyChips":["100k req/day","Edge runtime","KV + D1 + R2"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"cloudflare-workers","title":"Cloudflare Workers","provider":"Cloudflare Workers","planName":"freemium","logo":"/logos/cloudflare-workers-paid.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free tier historically centered around daily request quotas; verify current limits before launch.","keyMetrics":[{"key":"max_cpu_per_invocation","label":"CPU/call","value":"10 ms","display":"10 ms"},{"key":"bandwidth_per_month","label":"Bandwidth","value":"∞","display":"∞"}],"url":"https://developers.cloudflare.com/workers/"}},"nominations":[{"slug":"modal","label":"GPU compute","description":"Modal is the nomination for workloads that need GPU acceleration, Python containers, or recurring monthly compute credits rather than per-invocation pricing. The Starter plan includes $30 in free compute credits each month, which covers scheduled GPU inference jobs, batch processing, and sandboxed AI agent loops before the paid threshold arrives.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"modal","title":"Modal","provider":"Modal","planName":"freemium","logo":"/logos/modal.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Starter plan ($0): $30/month in free compute credits (recurring), 3 workspace seats, 100 containers + 10 GPU concurrency, 5 deployed crons, 8 webhook endpoints, 200 deployed apps, 1 TiB/month free volume storage.","keyMetrics":[{"key":"bandwidth_per_month","label":"Bandwidth","value":"∞","display":"∞"}],"url":"https://modal.com/pricing"}},{"slug":"upstash-workflow","label":"Durable workflow steps","description":"Upstash Workflow is the nomination for serverless functions that are really sequences of durable steps — read a document, call an LLM, write a result, send a notification. The free tier includes 1,000 steps per day and 15-minute maximum HTTP response duration, which covers lightweight agent loops without managing a queue service separately.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"upstash-workflow","title":"Upstash | Workflow","provider":"Upstash","planName":"freemium","logo":"/logos/upstash.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free tier: 1,000 steps/day, 1,000 steps per workflow run, 10 concurrent steps, 1 MB max message size, 7 day max sleep, and 15 min max HTTP response duration.","keyMetrics":[{"key":"bandwidth_per_month","label":"Bandwidth","value":"50 GB/mo","display":"50 GB/mo"}],"url":"https://upstash.com/docs/workflow"}}],"paidPicks":[]},"schedulers":{"collection":"infrastructure","category":"schedulers","updatedAt":"2026-07-10","intro":"Cron and background work services for recurring jobs, checks, queues, and workflows.","picks":[{"slug":"val-town-cron","headline":"The fastest cron for small scripts","description":"Val Town Cron is the hero scheduler pick because it turns a small scheduled script into a hosted recurring job with almost no setup. The free plan supports public vals and 15-minute cron intervals, 100,000 runs per day, and MCP integration — enough to cover most lightweight agent scheduling needs without configuring a pipeline or managing infrastructure. The wall-clock limit on the free plan keeps it honest for its weight class.","whyChips":["100k runs/day","15-min intervals","MCP integration"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"val-town-cron","title":"Val Town Cron","provider":"Val Town Cron","planName":"freemium","logo":"/logos/val-town.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free plan supports public vals and cron intervals; private vals and higher frequency require checking the current plan.","keyMetrics":[{"key":"cron_jobs","label":"Cron jobs","value":"2.5K jobs","display":"2.5K jobs"},{"key":"scheduler_workflow_runtime","label":"Workflow/runtime","value":"6 signals","display":"6 signals"},{"key":"min_interval","label":"Min interval","value":"15 min","display":"15 min"}],"url":"https://docs.val.town/vals/cron/"}},{"slug":"healthchecks-io","headline":"Dead-man switch for cron","description":"Healthchecks.io is the side pick for agent developers who want a simple dead-man switch: a service that alerts when a scheduled job fails to fire. The free plan includes 20 free checks, and the paid plan at roughly $16 per month gives 100 checks with longer log retention — making it a cheap insurance policy for agent workflows where missed runs have real consequences.","whyChips":["20 free checks","Alert on miss","Log retention"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"healthchecks-io","title":"Healthchecks.io","provider":"Healthchecks.io","planName":"paid","logo":"/logos/healthchecks-io.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Paid — $16/mo (Business, $192/yr). 100 checks (crons) at the Business plan ($192/yr, $16/mo) with longer log retention; 20 free checks; free Business plan for OSS/nonprofits. Alerts when jobs don't run on time.","keyMetrics":[{"key":"scheduler_run_quota","label":"Run quota","value":"1 runs/month","display":"1 runs/month"},{"key":"scheduler_workflow_runtime","label":"Workflow/runtime","value":"2 signals","display":"2 signals"}],"url":"https://healthchecks.io/pricing/"}},{"slug":"inngest","headline":"Durable workflows, not just cron","description":"Inngest earns its pick as the durable workflow scheduler for agent pipelines that go beyond simple cron: retries, branching, fan-out, and event-driven steps are first-class primitives rather than bolt-ons. The free Hobby plan includes 50,000 executions per month, 500k events per month, and unlimited serverless cron functions, with no card required. That quota handles meaningful agent workflow prototypes before scaling to a paid plan.","whyChips":["50k executions/mo","Durable retries","Unlimited cron"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"inngest","title":"Inngest","provider":"Inngest","planName":"freemium","logo":"/logos/inngest.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Hobby plan ($0/mo, no card): 50,000 executions/month, 500k events/month, 5 concurrent steps, 100k queue depth, unlimited serverless cron functions, and 24h trace/log retention.","keyMetrics":[{"key":"cron_jobs","label":"Cron jobs","value":"∞","display":"∞"},{"key":"scheduler_workflow_runtime","label":"Workflow/runtime","value":"12 signals","display":"12 signals"},{"key":"min_interval","label":"Min interval","value":"1 min","display":"1 min"}],"url":"https://www.inngest.com"}}],"hero":{"slug":"val-town-cron","headline":"The fastest cron for small scripts","description":"Val Town Cron is the hero scheduler pick because it turns a small scheduled script into a hosted recurring job with almost no setup. The free plan supports public vals and 15-minute cron intervals, 100,000 runs per day, and MCP integration — enough to cover most lightweight agent scheduling needs without configuring a pipeline or managing infrastructure. The wall-clock limit on the free plan keeps it honest for its weight class.","whyChips":["100k runs/day","15-min intervals","MCP integration"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"val-town-cron","title":"Val Town Cron","provider":"Val Town Cron","planName":"freemium","logo":"/logos/val-town.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free plan supports public vals and cron intervals; private vals and higher frequency require checking the current plan.","keyMetrics":[{"key":"cron_jobs","label":"Cron jobs","value":"2.5K jobs","display":"2.5K jobs"},{"key":"scheduler_workflow_runtime","label":"Workflow/runtime","value":"6 signals","display":"6 signals"},{"key":"min_interval","label":"Min interval","value":"15 min","display":"15 min"}],"url":"https://docs.val.town/vals/cron/"}},"nominations":[{"slug":"windmill","label":"Unlimited cron","description":"Windmill's Community Edition is self-hosted and completely free: unlimited cron schedules and executions across 3 workspaces with up to 50 users. It is the nomination for teams that want an open-source, code-first scheduler with a visual builder and execution history, and are comfortable operating the infrastructure themselves.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"windmill","title":"Windmill","provider":"Windmill","planName":"open_source_selfhost","logo":"/logos/paid-windmill.png","pricingModel":"open_source_selfhost","accessBucket":"conditional","freeLimitText":"Self-hosted Community Edition ($0): unlimited cron schedules and executions across 3 workspaces and up to 50 users (10 with SSO), with unlimited scripts/flows/apps and run-detail retention up to 30 days.","keyMetrics":[{"key":"scheduler_run_quota","label":"Run quota","value":"1 runs/month","display":"1 runs/month"},{"key":"cron_jobs","label":"Cron jobs","value":"∞","display":"∞"},{"key":"scheduler_workflow_runtime","label":"Workflow/runtime","value":"4 signals","display":"4 signals"}],"url":"https://www.windmill.dev"}},{"slug":"vercel-queues","label":"Most runs","description":"Vercel Queues includes the first 1 million Queue API operations per month free in the Hobby tier — a higher operational quota than most competing services at the free level. It is the nomination for agent developers already on Vercel who need to offload expensive work like emails, PDFs, and external API calls from synchronous request paths.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"vercel-queues","title":"Vercel Queues","provider":"Vercel Queues","planName":"freemium","logo":"/logos/vercel.svg","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Hobby includes the first 1,000,000 Queue API operations per month free; beyond that, regional on-demand pricing starting at $0.60 per 1M operations.","keyMetrics":[{"key":"executions","label":"API operations","value":"1M ops/mo","display":"1M ops/mo"},{"key":"scheduler_workflow_runtime","label":"Workflow/runtime","value":"9 signals","display":"9 signals"},{"key":"cron_jobs","label":"Topics","value":"∞","display":"∞"}],"url":"https://vercel.com/docs/queues"}}],"paidPicks":[]},"storage":{"collection":"infrastructure","category":"storage","updatedAt":"2026-07-10","intro":"Persistent, durable, bindable storage where free quota and egress economics matter.","picks":[{"slug":"cloudflare-r2-standard","headline":"The zero-egress object storage default","description":"Cloudflare R2 is the hero storage pick because it removes the cost developers fear most: egress to the public internet is free at every tier. The free tier includes 10 GB-months of storage, 1 million Class A operations and 10 million Class B operations per month. The S3-compatible API means existing tooling migrates without rewrite, and pairing with Cloudflare Workers keeps agent artifacts close to the compute that processes them.","whyChips":["$0 egress","10 GB storage","1M ops/mo write ops"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"cloudflare-r2-standard","title":"Cloudflare R2","provider":"Cloudflare R2","planName":"freemium","logo":"/logos/cloudflare-r2-infrequent-access.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free 10 GB stored + 1M Class A ops + 10M Class B ops each month, with ZERO egress fees and no minimum retention; beyond the free tier storage is $0.015/GB-month.","keyMetrics":[{"key":"egress_gb","label":"Egress","value":"$0 egress","display":"$0 egress"},{"key":"storage_gb","label":"Storage","value":"10 GB","display":"10 GB"},{"key":"class_a_ops","label":"Write ops","value":"1 M ops/mo","display":"1 M ops/mo"}],"url":"https://developers.cloudflare.com/r2/pricing/"}},{"slug":"backblaze-b2","headline":"Single cheap object storage","description":"Backblaze B2 is the plain-spoken storage pick: 10 GB free from the start, egress credits to reduce bills, and free egress up to 3x of stored data under the free-for-dev program. S3-compatible API, no credit card to create an account, and a straightforward pricing model make it a reliable baseline for backups, media assets, and agent output storage.","whyChips":["10 GB storage","3x free egress","S3-compatible"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"backblaze-b2","title":"Backblaze B2","provider":"Backblaze B2","planName":"freemium","logo":"/logos/backblaze-b2.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free 10 GB object storage for unlimited time (per free-for-dev); free egress up to 3x stored data; no credit card to create an account.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"10 GB","display":"10 GB"},{"key":"egress_gb","label":"Free egress","value":"30 GB/mo","display":"30 GB/mo"},{"key":"class_a_ops","label":"Write ops","value":"∞","display":"∞"}],"url":"https://www.backblaze.com/cloud-storage"}},{"slug":"storj","headline":"Distributed storage with a no-card path","description":"Storj is the storage pick for applications where decentralized durability and S3-compatible storage are part of the value proposition. Accounts without a payment method get 25 GB of storage and 25 GB of egress per month with no monthly fee and no credit card required. The trade-off is that the free tier uses shared infrastructure, so it suits non-critical assets and archival more than hot production data.","whyChips":["25 GB storage","25 GB egress","No card"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"storj","title":"Storj","provider":"Storj","planName":"freemium","logo":"/logos/paid-storj.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Accounts without a payment method get 25 GB of storage and 25 GB of egress per month with no monthly fee and no credit card required.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"25 GB","display":"25 GB"},{"key":"egress_gb","label":"Free egress","value":"25 GB/mo","display":"25 GB/mo"}],"url":"https://www.storj.io"}}],"hero":{"slug":"cloudflare-r2-standard","headline":"The zero-egress object storage default","description":"Cloudflare R2 is the hero storage pick because it removes the cost developers fear most: egress to the public internet is free at every tier. The free tier includes 10 GB-months of storage, 1 million Class A operations and 10 million Class B operations per month. The S3-compatible API means existing tooling migrates without rewrite, and pairing with Cloudflare Workers keeps agent artifacts close to the compute that processes them.","whyChips":["$0 egress","10 GB storage","1M ops/mo write ops"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"cloudflare-r2-standard","title":"Cloudflare R2","provider":"Cloudflare R2","planName":"freemium","logo":"/logos/cloudflare-r2-infrequent-access.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free 10 GB stored + 1M Class A ops + 10M Class B ops each month, with ZERO egress fees and no minimum retention; beyond the free tier storage is $0.015/GB-month.","keyMetrics":[{"key":"egress_gb","label":"Egress","value":"$0 egress","display":"$0 egress"},{"key":"storage_gb","label":"Storage","value":"10 GB","display":"10 GB"},{"key":"class_a_ops","label":"Write ops","value":"1 M ops/mo","display":"1 M ops/mo"}],"url":"https://developers.cloudflare.com/r2/pricing/"}},"nominations":[{"slug":"tigris","label":"Zero egress","description":"Tigris gives S3-standard storage with 5 GB per month free, 10,000 Class A and 100,000 Class B requests free, and zero egress fees across all tiers. It is a clean pick when you want S3 compatibility without bandwidth bills, especially for data-intensive agent workloads that read more than they write.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"tigris","title":"Tigris Data","provider":"Tigris Data","planName":"freemium","logo":"/logos/tigris.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"5GB of Standard storage/month free, 10,000 Class A requests/month free, 100,000 Class B requests/month free, and zero egress fees on all tiers.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"5 GB","display":"5 GB"},{"key":"egress_gb","label":"Free egress","value":"∞","display":"∞"},{"key":"class_b_ops","label":"Read ops","value":"0.1 M ops/mo","display":"0.1 M ops/mo"}],"url":"https://www.tigrisdata.com/pricing/"}},{"slug":"telnyx-cloud-storage","label":"Most free GB","description":"Telnyx Cloud Storage includes 10 GiB of storage, 1 million state-change write operations, and 10 million state-read operations per month, all with no egress fees. The quota is unusually generous among the conditional-free tiers, making it a practical choice for agent apps that accumulate media or document assets.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"telnyx-cloud-storage","title":"Telnyx Cloud Storage","provider":"Telnyx Cloud Storage","planName":"freemium","logo":"/logos/paid-telnyx-cloud-storage.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"10 GiB of storage, 1 million state-change (write) operations, and 10 million state-read operations per month, with no egress fees.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"10 GB","display":"10 GB"},{"key":"egress_gb","label":"Free egress","value":"∞","display":"∞"},{"key":"class_a_ops","label":"Write ops","value":"1 M ops/mo","display":"1 M ops/mo"}],"url":"https://telnyx.com/products/cloud-storage"}}],"paidPicks":[]},"workspaces":{"collection":"infrastructure","category":"workspaces","updatedAt":"2026-07-10","intro":"Team workspaces, docs, and knowledge tools with free plans for agent project management.","picks":[{"slug":"grist","headline":"The relational database workspace","description":"Grist is the hero workspace pick because it is a spreadsheet-database hybrid that non-engineers can use fluently and engineers can extend with Python. The Cloud Free plan allows unlimited members and 2 guests per doc with 1 GB attachment storage per doc, and self-hosting is free for individuals and orgs with under $1M annual income. That combination makes it a durable, self-hostable foundation for agent-run projects where the output is structured data rather than documents.","whyChips":["Unlimited members","1 GB attachments","Self-hostable"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"grist","title":"Grist","provider":"Grist","planName":"freemium","logo":"/logos/grist.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Cloud Free plan: unlimited members, 2 guests per doc, 1 GB attachment storage per doc. Self-hosting is $0 for individuals and orgs with under $1M total annual income (free activation key).","keyMetrics":[{"key":"workspace_content_api_access","label":"Content","value":"4 level","display":"4 level"},{"key":"workspace_collaboration_seats","label":"Seats","value":"5 level","display":"5 level"},{"key":"workspace_automation_integration_readiness","label":"Auto","value":"4 level","display":"4 level"}],"url":"https://www.getgrist.com/"}},{"slug":"clickup","headline":"Best broad free team workspace","description":"ClickUp is the side pick because its Free Forever plan covers tasks, docs, whiteboards, and common project views in one place. It is broad enough to act as a lightweight operating system for a small project team, with collaborative docs, list and board views, and real-time collaboration all included without a credit card. The storage cap of 60 MB is the honest trade-off.","whyChips":["Tasks + docs","Unlimited members","No card"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"clickup","title":"ClickUp","provider":"ClickUp","planName":"freemium","logo":"/logos/clickup-business.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free Forever plan: unlimited tasks and unlimited free-plan members with collaborative Docs, whiteboards, and list/board/calendar views, capped at 60 MB of file storage, 5 Spaces, and 100 uses of advanced features.","keyMetrics":[{"key":"workspace_content_api_access","label":"Content","value":"4 level","display":"4 level"},{"key":"workspace_collaboration_seats","label":"Seats","value":"5 level","display":"5 level"},{"key":"workspace_automation_integration_readiness","label":"Auto","value":"3 level","display":"3 level"}],"url":"https://clickup.com"}},{"slug":"nuclino","headline":"Fast wiki-style team knowledge","description":"Nuclino is the side pick for teams that want a fast, no-clutter shared wiki with real-time collaboration and visual graph navigation, all free for unlimited users with up to 50 items and 2 GB total storage. The hard item cap of 50 means it fits small teams and focused projects rather than large knowledge bases, but for agent-assisted research and documentation it is a clean tool.","whyChips":["Unlimited users","2 GB storage","Real-time collab"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"nuclino","title":"Nuclino","provider":"Nuclino","planName":"freemium","logo":"/logos/nuclino-business.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free forever for unlimited users with up to 50 items (docs/pages) and 2 GB total storage, including real-time collaboration, all views, and 30-day version history.","keyMetrics":[{"key":"workspace_content_api_access","label":"Content","value":"4 level","display":"4 level"},{"key":"workspace_collaboration_seats","label":"Seats","value":"5 level","display":"5 level"}],"url":"https://www.nuclino.com"}}],"hero":{"slug":"grist","headline":"The relational database workspace","description":"Grist is the hero workspace pick because it is a spreadsheet-database hybrid that non-engineers can use fluently and engineers can extend with Python. The Cloud Free plan allows unlimited members and 2 guests per doc with 1 GB attachment storage per doc, and self-hosting is free for individuals and orgs with under $1M annual income. That combination makes it a durable, self-hostable foundation for agent-run projects where the output is structured data rather than documents.","whyChips":["Unlimited members","1 GB attachments","Self-hostable"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"grist","title":"Grist","provider":"Grist","planName":"freemium","logo":"/logos/grist.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Cloud Free plan: unlimited members, 2 guests per doc, 1 GB attachment storage per doc. Self-hosting is $0 for individuals and orgs with under $1M total annual income (free activation key).","keyMetrics":[{"key":"workspace_content_api_access","label":"Content","value":"4 level","display":"4 level"},{"key":"workspace_collaboration_seats","label":"Seats","value":"5 level","display":"5 level"},{"key":"workspace_automation_integration_readiness","label":"Auto","value":"4 level","display":"4 level"}],"url":"https://www.getgrist.com/"}},"nominations":[{"slug":"cryptpad","label":"Encrypted","description":"CryptPad is the nomination for teams that need end-to-end encryption as a first-class property — the server and operator cannot read stored documents. A free registered account gives 1 GB of encrypted storage across all apps (docs, sheets, slides, code, kanban, forms, whiteboard, calendar) with real-time collaboration. Per-file uploads are capped at 25 MB on the free plan.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"cryptpad","title":"CryptPad","provider":"CryptPad","planName":"freemium","logo":"/logos/cryptpad.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free registered account on cryptpad.fr gives 1 GB of end-to-end-encrypted storage with full access to every app (docs, sheets, slides, code/markdown, kanban, forms, whiteboard, calendar), real-time collaboration, teams, and shared folders, with per-file uploads capped at 25 MB.","keyMetrics":[{"key":"workspace_content_api_access","label":"Content","value":"3 level","display":"3 level"},{"key":"workspace_collaboration_seats","label":"Seats","value":"1 level","display":"1 level"}],"url":"https://cryptpad.fr/"}},{"slug":"affine","label":"Open-source","description":"AFFiNE is the nomination for teams that want a FOSS-leaning, local-first workspace with a Notion + Miro hybrid feature set. The free cloud plan includes unlimited local workspaces, 10 GB cloud storage, and up to 3 members per workspace, and the MIT-licensed FOSS edition is fully self-hostable — which means agent data never needs to leave the team's own infrastructure.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"affine","title":"AFFiNE","provider":"AFFiNE","planName":"freemium","logo":"/logos/affine.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free forever: unlimited local workspaces, 10 GB cloud storage, up to 3 members per workspace, 7-day cloud version history. FOSS edition fully self-hostable under MIT.","keyMetrics":[{"key":"workspace_content_api_access","label":"Content","value":"3 level","display":"3 level"},{"key":"workspace_collaboration_seats","label":"Seats","value":"2 level","display":"2 level"}],"url":"https://affine.pro/"}}],"paidPicks":[]},"domains":{"collection":"infrastructure","category":"domains","updatedAt":"2026-07-10","intro":"Free DNS, dynamic DNS, and provider subdomain options for agent-accessible hostnames.","picks":[{"slug":"duckdns","headline":"Simple dynamic DNS for free subdomains","description":"Duck DNS is the hero domain pick because it is the lowest-friction way to get a public hostname that tracks a changing home or office IP address. Token-based updates mean the update script can live in a router, a small device, or an agent cron without storing credentials in plaintext. The trade-off is that you only get *.duckdns.org subdomains — no custom apex domains — and login options have narrowed over time.","whyChips":["No card","Token updates","Free subdomain"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"duckdns","title":"Duck DNS","provider":"Duck DNS","planName":"free_forever","logo":"/logos/duckdns.png","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Free: multiple *.duckdns.org subdomains with token-based dynamic IP updates.","keyMetrics":[{"key":"free_domains","label":"Free domains","value":"5 domains","display":"5 domains"},{"key":"ddns_update_interval","label":"DDNS refresh","value":"5 min","display":"5 min"}],"url":"https://www.duckdns.org/"}},{"slug":"cloudns","headline":"Feature-rich free DNS","description":"ClouDNS is the side pick for developers who want more than basic A records from their free DNS provider. The Free Forever plan includes 1 DNS zone, 50 DNS records, 500,000 DNS queries per month on 4 unicast nameservers, plus 1 mail forward and 1 dynamic DNS hostname. That quota handles personal projects and small services without immediately hitting limits.","whyChips":["1 zone free","500k queries/mo","Dynamic DNS"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"cloudns","title":"ClouDNS","provider":"ClouDNS","planName":"freemium","logo":"/logos/cloudns-ddos-protected-s.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free Forever: 1 DNS zone, 50 DNS records, 500,000 DNS queries/month on 4 unicast nameservers, plus 1 mail forward and 1 dynamic DNS hostname.","keyMetrics":[{"key":"free_domains","label":"Free domains","value":"1 domains","display":"1 domains"}],"url":"https://www.cloudns.net/"}},{"slug":"dynv6","headline":"Unlimited free hosts with DNSSEC","description":"dynv6 is the side pick for self-hosters and home-lab users who want free public hostnames under dynv6.net, dns.army, or v6.rocks with full DNS record management and DNSSEC at no cost. The service is entirely free for private use with unlimited hosts, which makes it practical for agent fleets and IoT setups where each instance needs its own resolvable hostname.","whyChips":["Unlimited hosts","DNSSEC","No card"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"dynv6","title":"dynv6","provider":"dynv6","planName":"free_forever","logo":"/logos/dynv6.png","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Completely free for private use: unlimited hosts with freely chosen names under dynv6.net/dns.army/v6.rocks etc. (or delegate your own domain), with full DNS record management and DNSSEC at no cost.","keyMetrics":[{"key":"free_domains","label":"Free domains","value":"∞","display":"∞"}],"url":"https://dynv6.com/"}}],"hero":{"slug":"duckdns","headline":"Simple dynamic DNS for free subdomains","description":"Duck DNS is the hero domain pick because it is the lowest-friction way to get a public hostname that tracks a changing home or office IP address. Token-based updates mean the update script can live in a router, a small device, or an agent cron without storing credentials in plaintext. The trade-off is that you only get *.duckdns.org subdomains — no custom apex domains — and login options have narrowed over time.","whyChips":["No card","Token updates","Free subdomain"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"duckdns","title":"Duck DNS","provider":"Duck DNS","planName":"free_forever","logo":"/logos/duckdns.png","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Free: multiple *.duckdns.org subdomains with token-based dynamic IP updates.","keyMetrics":[{"key":"free_domains","label":"Free domains","value":"5 domains","display":"5 domains"},{"key":"ddns_update_interval","label":"DDNS refresh","value":"5 min","display":"5 min"}],"url":"https://www.duckdns.org/"}},"nominations":[{"slug":"sslip-io","label":"No signup","description":"sslip.io resolves any IP-embedded hostname instantly with no account and no credit card — for example, 192-168-0-1.sslip.io resolves to 192.168.0.1. It is the nomination for developers who need valid hostnames with built-in TLS certificates for local servers, Kubernetes ingress, and demos without registering a domain at all.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"sslip-io","title":"sslip.io","provider":"sslip.io","planName":"free_forever","logo":"/logos/sslip-io.png","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Unlimited resolution of any IPv4/IPv6-embedded hostname (e.g. 192-168-0-1.sslip.io) at no cost, with no account and no card; optional self-hosted branded subdomains via NS delegation.","keyMetrics":[{"key":"free_domains","label":"Free domains","value":"∞","display":"∞"}],"url":"https://sslip.io/"}},{"slug":"pages-dev","label":"Free subdomains","description":"Cloudflare pages.dev is the nomination for frontend-focused agent deployments where the agent's output is a static site or web app. It provides a free *.pages.dev subdomain with automatic TLS and global CDN, requiring nothing more than a GitHub repository connection.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"pages-dev","title":"Cloudflare pages.dev","provider":"Cloudflare pages.dev","planName":"free_forever","logo":"/logos/cloudflare.svg","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Free provider subdomain.","keyMetrics":[{"key":"free_domains","label":"Free domains","value":"∞","display":"∞"},{"key":"dns_records","label":"DNS records","value":"∞","display":"∞"}],"url":"https://developers.cloudflare.com/pages/"}}],"paidPicks":[]}},"partial":true}