# DeepSeek-R1 API prices

DeepSeek-R1 (deepseek-thinking) — 164k context, 41k max output.
13 metered per-token offers from 15 provider(s), in USD per 1M tokens.

| Provider | Input $/1M | Output $/1M | Cache read $/1M | Context |
| --- | --- | --- | --- | --- |
| NanoGPT | $0.4 | $1.7 | $0.2 | 128k |
| D.Run (China) | $0.55 | $2.2 | — | 131k |
| Alibaba (China) | $0.574 | $2.294 | — | 131k |
| Eden AI | $0.7 | $2.4 | — | 164k |
| Hugging Face | $0.7 | $2.5 | — | 64k |
| Kilo Gateway | $0.7 | $2.5 | — | 64k |
| OpenRouter | $0.7 | $2.5 | — | 64k |
| Amazon Bedrock | $1.35 | $5.4 | — | 128k |
| Amazon Bedrock (us) | $1.35 | $5.4 | — | 128k |
| Merge Gateway | $1.35 | $5.4 | — | 164k |
| Vercel AI Gateway | $1.35 | $5.4 | — | 128k |
| IO.NET | $2 | $8.75 | $1 | 128k |
| Abacus | $3 | $7 | — | 128k |

## Summary

- Cheapest input: $0.4 per 1M tokens (NanoGPT)
- Cheapest output: $1.7 per 1M tokens (NanoGPT)
- First-party: not listed separately
- Free offers: iFlow
- Subscription plans (not per-token): none
- Released: 2025-01-20
- Inputs: text

HTML page: https://llmprice.gitlab.io/models/deepseek-r1/
Full dataset: https://llmprice.gitlab.io/data/catalog.json