# DeepSeek V4.1 Flash 1017

DeepSeek V4.1 Flash 1017 is a third-party open weight model from DeepSeek, hosted by Mistral, with support for text and image inputs. Ideal for long-context coding and agentic workflows. The model is served without Mistral modifications.

- API names: `deepseek-v4-1-flash-1017`
- Status: Public Preview
- Type: Open, Generalist
- Release date: 2026-10-28
- Version: v4.1-flash-1017
- Context length: 1M
- Output token limit: 128k
- Input: text, image
- Output: reasoning, text
- Features: structured-outputs, function-calling, prefix, chat-completions, batching, agents-conversations, connectors
- Speed rating: 5/5
- Performance rating: 5/5
- Pricing: Input $0.3 /M Tokens; Cached input $0.007 /M Tokens; Output $1.2 /M Tokens
- Licenses: MIT
- Open weights: [DeepSeek V4.1 Flash 1017](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017), 763B parameters, 37B active, MIT
