PROVIDER PRICING
Replicate API Pricing
Replicate API pricing: rates for 20 models, how they compare to official pricing, and what it costs to run.
MODEL PRICING
API model pricing
20 models · prices per million tokens
| Model | Input / 1M | Output / 1M |
|---|---|---|
| cpu | — | — |
| cpu-small | — | — |
| gpu-a100-large | — | — |
| gpu-a100-large-8x | — | — |
| gpu-h100 | — | — |
| gpu-h100-2x | — | — |
| gpu-h100-4x | — | — |
| gpu-h100-8x | — | — |
| gpu-l40s | — | — |
| gpu-l40s-8x | — | — |
| gpu-t4 | — | — |
Prices change frequently; confirm current rates on the provider’s official pricing page.