Hugging Face Inference
LLM 4.0/10Model access through Hugging Face ecosystem and providers.
Hugging Face Inference scores 4/10 for free value in LLM. The main constraint: Free/provider credits and limits vary by account, provider, and model.
Free value score
4.0 / 10Show the math
Key metrics
Models
40 free · 125+ total| Model | Family | AA | Access |
|---|---|---|---|
| zai-org/GLM-5.2 | GLM | 51.1 | free |
| deepseek-ai/DeepSeek-V4-Pro | DeepSeek | 44.3 | free |
| moonshotai/Kimi-K2.6 | Kimi | 42.8 | free |
| moonshotai/Kimi-K2.7-Code | Kimi | 42 | free |
| deepseek-ai/DeepSeek-V4-Flash | DeepSeek | 40.3 | free |
| nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4 | Nemotron | 37.8 | free |
| Qwen/Qwen3.5-397B-A17B | Qwen | 32 | free |
| deepseek-ai/DeepSeek-V3.2-Exp | DeepSeek | 24.7 | free |
| openai/gpt-oss-120b | GPT | 23.8 | free |
| zai-org/GLM-4.6 | GLM | 23 | free |
| deepseek-ai/DeepSeek-V3.1-Terminus | DeepSeek | 21.1 | free |
| deepseek-ai/DeepSeek-R1 | DeepSeek | 20.1 | free |
| moonshotai/Kimi-K2-Instruct | Kimi | 19.4 | free |
| moonshotai/Kimi-K2-Thinking | Kimi | 19.4 | free |
| zai-org/GLM-4.5-Air | GLM | 16.5 | free |
| openai/gpt-oss-20b | GPT | 14.9 | free |
| meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8 | Llama | 14.3 | free |
| meta-llama/Llama-4-Scout-17B-16E-Instruct | Llama | 10 | free |
| Qwen/Qwen2.5-Coder-32B-Instruct | Qwen | 7.1 | free |
| CohereLabs/c4ai-command-a-03-2025 | Cohere | — | free |
| CohereLabs/command-a-reasoning-08-2025 | Cohere | — | free |
| deepseek-ai/DeepSeek-Prover-V2-671B | DeepSeek | — | free |
| deepseek-ai/DeepSeek-R1-0528 | DeepSeek | — | free |
| deepseek-ai/DeepSeek-V3-0324 | DeepSeek | — | free |
| google/gemma-3-27b-it | Gemma | — | free |
| meta-llama/Llama-3.3-70B-Instruct | Llama | — | free |
| MiniMaxAI/MiniMax-M2 | MiniMax | — | free |
| MiniMaxAI/MiniMax-M2.5 | MiniMax | — | free |
| mistralai-not-listed/placeholder | Mistral | — | free |
| moonshotai/Kimi-K2-Instruct-0905 | Kimi | — | free |
| openai/gpt-oss-safeguard-20b | GPT | — | free |
| Qwen/Qwen2.5-72B-Instruct | Qwen | — | free |
| Qwen/Qwen3-235B-A22B-Instruct-2507 | Qwen | — | free |
| Qwen/Qwen3-235B-A22B-Thinking-2507 | Qwen | — | free |
| Qwen/Qwen3-Coder-480B-A35B-Instruct | Qwen | — | free |
| Qwen/Qwen3-Next-80B-A3B-Thinking | Qwen | — | free |
| Qwen/Qwen3-VL-235B-A22B-Instruct | Qwen | — | free |
| Qwen/QwQ-32B | Qwen | — | free |
| zai-org/GLM-4.5 | GLM | — | free |
| zai-org/GLM-4.6V | GLM | — | free |
Showing first 40 of a larger catalog.
What you get
Provider marketplace
Free credits vary
Good with Spaces/Datasets
Model/provider-specific limits
Limitations
- Card required for pay-as-you-go beyond credits
Continuing inference past the monthly $0.10 credit requires purchasing additional credits, which requires a payment method on file.
- Free credits subject to removal
HF documentation explicitly states the $0.10 monthly free credits are 'subject to change', meaning they can be reduced or eliminated at any time.
Sources
-
"Free Users | $0.10, subject to change | yes (credits purchase required)"
https://huggingface.co/docs/inference-providers/en/pricing -
"Every Hugging Face user receives monthly credits to experiment with Inference Providers"
https://huggingface.co/docs/inference-providers/en/pricing -
"@Kimi_Moonshot's Kimi K2 Thinking achieves a 67 in the Artificial Analysis Intelligence Index. This positions it clearly above all other open[-weights models]"
https://x.com/ArtificialAnlys/status/1986911675820446013 -
"Hugging Face grew to 13 million users, more than 2 million public models, and over 500,000 public datasets"
https://huggingface.co/blog/huggingface/state-of-os-hf-spring-2026