Replicate

Functions 4.6/10

Pay-per-second by hardware: CPU $0.000025/s, T4 $0.000225/s ($0.81/hr), L40S $3.51/hr, A100 80GB $5.04/hr, H100 $5.49/hr; multi-GPU available. Public models billed only for active

Replicate scores 4.6/10 for free value in Functions. The main constraint: Paid — T4 $0.000225/s; A100 80GB $0.0014/s; H100 $0.001525/s. Pay-per-second by hardware: CPU $0.000025/s, T4 $0.000225/s ($0.81/hr), L40S $3.51/hr, A100 80GB $5.04/hr, H100 $5.49/hr; multi-GPU available. Public models billed only for active time.

Category Functions Kind Paid plan

Free value score

4.6 / 10

Show the math

Overall 4.6/10

value = 4.6/100

weighted_components_v1

Invocation value 1/10

value = 0.01 ×

fixed_log_v1

Runtime fit 8.2/10

value = 8.2/100

policy_score_v1

Deploy friction 8/10

value = 8/100

policy_score_v1

Key metrics

T4 $.81/hr

A100 $1.4m/s

H100 $1.5m

What you get

Paid — T4 $0.000225/s; A100 80GB $0.0014/s; H100 $0.001525/s. Pay-per-second by hardware: CPU $0.000025/s, T4 $0.000225/s ($0.81/hr), L40S $3.51/hr, A100 80GB $5.04/hr, H100 $5.49/hr; multi-GPU available. Public models billed only for active time.

T4 at $0.81/hr is cheap for light inference; A100/H100 rates higher than RunPod/Modal because private-model billing includes cold-start/idle. Best val

Sources

"Nvidia A100 (80GB) GPU gpu-a100-large $0.001400/sec $5.04/hr ... Nvidia H100 GPU gpu-h100 $0.001525/sec $5.49/hr"
https://replicate.com/pricing
"Nvidia T4 GPU gpu-t4 $0.000225/sec $0.81/hr"
https://replicate.com/pricing

paid value serverless-functions

Open service ← All Functions