# Mistral DeepSeek V4.1 Flash 1017 API price

Mistral DeepSeek V4.1 Flash 1017: input $0.30, cached input $0.007, output $1.20 per 1M tokens (Standard).

- Input: $0.30 per 1M tokens
- Cached input: $0.007 per 1M tokens
- Output: $1.20 per 1M tokens
- Rate label: Standard
- Context window: 1M tokens
- Max output: 128k tokens
- Modalities: text, image → text
- Status: Preview
- Release date: 2026-10-28
- API id: deepseek-v4-1-flash-1017
- Weights from: DeepSeek

Third-party open-weight model from DeepSeek (MIT), hosted by Mistral without modifications. The Mistral model card lists Public Preview, a 1M context window, 128k max output, text and image input, and $0.30 input, $0.007 cached input, $1.20 output per 1M tokens; the same rates are on the Mistral pricing page. This is the same model version DeepSeek serves as deepseek-flash on its own API, where it is billed at different peak and off-peak rates.

## Sources

- Pricing: https://docs.mistral.ai/models/deepseek-v4-flash-1017
- Limits: https://docs.mistral.ai/models/deepseek-v4-flash-1017
- Release: https://docs.mistral.ai/models/deepseek-v4-flash-1017

Checked: 2026-11-01

Canonical page: https://modelmeter.live/models/deepseek-v4-1-flash-1017

Dataset: https://modelmeter.live/api/models.json

## Price changes

- 2026-11-01 new model: n/a -> $0.30 / $0.007 / $1.20 per 1M tokens (input / cached input / output). DeepSeek V4.1 Flash 1017 (deepseek-v4-1-flash-1017), a DeepSeek open-weight model hosted by Mistral without modifications, is $0.30 input, $0.007 cached input, and $1.20 output per 1M tokens; the Mistral model card lists Public Preview and a release date of 28 October 2026, and this row was added to the ledger on 1 November 2026. Source: https://docs.mistral.ai/models/deepseek-v4-flash-1017