---
id: "deepseek/deepseek-v4.1-flash-1017"
also: ["deepseek/deepseek-v4.1-flash-20261017", "deepseek-ai/DeepSeek-V4.1-Flash-1017", "hf/deepseek-ai/DeepSeek-V4.1-Flash-1017", "requesty/deepseek/deepseek-v4.1-flash-1017", "vercel/deepseek/deepseek-v4.1-flash-1017"]
card_usd_per_m: "0.3 in, 1.2 out, at deepseek"
cheapest_list_usd_per_m: "0.006 in, 0.5 out, at relace/fp4 on openrouter"
cheapest_effective_usd_per_m: "0.0044 in, 0.2353 out, at deepinfra/fp8/flex on openrouter"
context: 1048576
max_output: 943718
input: ["text", "image"]
output: ["text"]
released: "2026-10-17"
reasoning: ["max", "high", "low", "none"]
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["frequency_penalty", "include_reasoning", "logit_bias", "logprobs", "max_completion_tokens", "max_tokens", "min_p", "parallel_tool_calls", "presence_penalty", "reasoning", "reasoning_effort", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_logprobs", "top_p"]
endpoints: 86
gateways: "openrouter (52), vercel (24), requesty (8), hf (2)"
no_training: 73
trains: 9
retains_prompts: 18
intelligence: 28.6000
---
# DeepSeek: DeepSeek V4.1 Flash 1017

DeepSeek V4.1 Flash 1017 is a minor update to DeepSeek V4.1 Flash with substantial gains in reasoning, coding, agentic tasks and visual benchmarks, plus shorter outputs and faster task completion. 552B parameter Mixture of Experts with native vision, 1M token context, thinking and non thinking modes, tool calling, structured JSON output and prompt caching. Served on DeepInfra in FP8 with zero retention. Open weights on Hugging Face.