---
id: "deepseek/deepseek-v4-flash-0731"
also: ["deepseek/deepseek-v4-flash-20260731", "deepseek-ai/DeepSeek-V4-Flash-0731", "hf/deepseek-ai/DeepSeek-V4-Flash-0731", "requesty/deepseek/deepseek-v4-flash-0731", "vercel/deepseek/deepseek-v4-flash-0731"]
card_usd_per_m: "0.44 in, 1.32 out, at deepseek"
cheapest_list_usd_per_m: "0.03 in, 0.1 out, at open-inference/fp4 on openrouter"
cheapest_effective_usd_per_m: "0.0143 in, 0.3956 out, at streamlake/fp8 on openrouter"
context: 1048576
max_output: 943718
input: ["text"]
output: ["text"]
released: "2026-07-31"
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_tokens", "min_p", "parallel_tool_calls", "presence_penalty", "reasoning", "reasoning_effort", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_a", "top_k", "top_logprobs", "top_p"]
endpoints: 91
gateways: "openrouter (47), vercel (26), requesty (11), hf (7)"
no_training: 66
trains: 8
retains_prompts: 20
coding: 69.1000
agentic: 41
---
# DeepSeek: DeepSeek V4 Flash 0731

DeepSeek V4 Flash 0731 is the official release of DeepSeek V4 Flash, superseding the preview version, with substantially enhanced agentic capabilities. A 304B parameter open-source MoE model with a 1M token context window, tuned for fast inference and high-throughput use cases with strong reasoning, coding, and function calling performance.