# DeepSeek V3.1 Terminus API prices

DeepSeek V3.1 Terminus (deepseek) — 164k context, 66k max output.
8 metered per-token offers from 8 provider(s), in USD per 1M tokens.

| Provider | Input $/1M | Output $/1M | Cache read $/1M | Context |
| --- | --- | --- | --- | --- |
| NanoGPT | $0.25 | $0.7 | $0.125 | 128k |
| NanoGPT | $0.25 | $0.7 | $0.125 | 128k |
| Abacus | $0.27 | $1 | — | 128k |
| Helicone | $0.27 | $1 | $0.216 | 128k |
| Kilo Gateway | $0.27 | $1 | $0.135 | 131k |
| NovitaAI | $0.27 | $1 | $0.135 | 131k |
| OpenRouter | $0.27 | $1 | $0.135 | 164k |
| Vercel AI Gateway | $0.27 | $1 | $0.135 | 131k |

## Summary

- Cheapest input: $0.25 per 1M tokens (NanoGPT)
- Cheapest output: $0.7 per 1M tokens (NanoGPT)
- First-party: not listed separately
- Free offers: none
- Subscription plans (not per-token): none
- Released: 2025-09-22
- Inputs: text

HTML page: https://llmprice.gitlab.io/models/deepseek-v3-1-terminus/
Full dataset: https://llmprice.gitlab.io/data/catalog.json