# DeepSeek V3.2 Thinking API prices

DeepSeek V3.2 Thinking (deepseek) — 163k context, 128k max output.
3 metered per-token offers from 3 provider(s), in USD per 1M tokens.

| Provider | Input $/1M | Output $/1M | Cache read $/1M | Context |
| --- | --- | --- | --- | --- |
| NanoGPT | $0.28 | $0.42 | $0.14 | 163k |
| 302.AI | $0.29 | $0.43 | — | 128k |
| Vercel AI Gateway | $0.62 | $1.85 | — | 128k |

## Summary

- Cheapest input: $0.28 per 1M tokens (NanoGPT)
- Cheapest output: $0.42 per 1M tokens (NanoGPT)
- First-party: not listed separately
- Free offers: none
- Subscription plans (not per-token): none
- Released: 2025-12-01
- Inputs: pdf, text

HTML page: https://llmprice.gitlab.io/models/deepseek-v3-2-thinking/
Full dataset: https://llmprice.gitlab.io/data/catalog.json