---
id: "deepseek/deepseek-v4-flash-vision-exp"
also: ["deepseek/deepseek-v4-flash-vision-exp-20260821", "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp", "hf/deepseek-ai/DeepSeek-V4-Flash-Vision-Exp", "requesty/deepseek/deepseek-v4-flash-vision-exp", "vercel/deepseek/deepseek-v4-flash-vision-exp"]
card_usd_per_m: "0.44 in, 1.32 out, at deepseek"
cheapest_list_usd_per_m: "0.2156 in, 0.6468 out, at deepinfra/fp8 on openrouter"
cheapest_effective_usd_per_m: "0.019 in, 0.66 out, at fireworks on requesty"
context: 1048576
max_output: 943717
input: ["text", "image"]
output: ["text"]
released: "2026-08-21"
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", "presence_penalty", "reasoning", "reasoning_effort", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_logprobs", "top_p"]
endpoints: 16
gateways: "openrouter (7), vercel (5), hf (3), requesty (1)"
no_training: 10
trains: 1
retains_prompts: 4
coding: 65
agentic: 47.5000
---
# DeepSeek: DeepSeek V4 Flash Vision Exp

DeepSeek-V4-Flash-Vision-Exp is an experimental multimodal model that combines the agentic reasoning, coding, and world knowledge capabilities of DeepSeek-V4-Flash with advanced visual understanding. It delivers major gains on vision-dependent agent workflows while performing strongly across leading coding, repository, data science, automation, and multimodal benchmarks.