Back to Models

DeepSeek V4 Flash

deepseekdeepseek/deepseek-v4-flash

How TheRouter serves this differently from the vendor

As the vendor operates it

DeepSeek's first-party service exposes V4-Flash as deepseek-v4-flash through OpenAI-format, Anthropic-format, and Responses API endpoints; Hugging Face also publishes the open weights for self-hosting.

On TheRouter

TheRouter serves the same text model under deepseek/deepseek-v4-flash on its OpenAI-compatible /v1 surface. Pricing, context length, output limit, and supported parameters are rendered from the live model registry rather than duplicated in this curated page.

API guide

Chat completion

Standard OpenAI-compatible chat endpoint. Add reasoning_effort and the thinking extra_body to enable chain-of-thought.

cURL
curl https://api.therouter.ai/v1/chat/completions \
  -H "Authorization: Bearer $THEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek/deepseek-v4-flash",
    "messages": [{"role":"user","content":"Explain quantum computing in simple terms"}],
    "reasoning_effort": "high",
    "extra_body": {"thinking": {"type": "enabled"}}
  }'

Thinking / reasoning mode

V4-Flash supports explicit thinking with separate reasoning_content output. Use high or max effort for hard agentic tasks.

cURL
curl https://api.therouter.ai/v1/chat/completions \
  -H "Authorization: Bearer $THEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek/deepseek-v4-flash",
    "messages": [{"role":"user","content":"Solve this step by step: ..."}],
    "reasoning_effort": "max",
    "extra_body": {"thinking": {"type": "enabled"}}
  }'

Tool calling

cURL
curl https://api.therouter.ai/v1/chat/completions \
  -H "Authorization: Bearer $THEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek/deepseek-v4-flash",
    "messages": [{"role":"user","content":"What's the weather in SF?"}],
    "tools": [{"type":"function","function":{"name":"get_weather","parameters":{...}}}],
    "tool_choice": "auto"
  }'

Related news

Guides

Fact ledger β€” every claim on this page traces here
sourceURLretrieved
Release date (preview)api-docs.deepseek.com β†—2026-05-25verified
Architectureapi-docs.deepseek.com β†—2026-05-25verified
Context windowapi-docs.deepseek.com β†—2026-05-25verified
Max outputapi-docs.deepseek.com β†—2026-05-25verified
Reasoning modesapi-docs.deepseek.com β†—2026-05-25verified
License β€” codehuggingface.co β†—2026-05-25verified
License β€” weightshuggingface.co β†—2026-05-25verified
Open weightshuggingface.co β†—2026-05-25verified
Training tokenshuggingface.co β†—2026-08-12verified
Terminal Bench 2.0huggingface.co β†—2026-08-12verified
LiveCodeBench Pass@1huggingface.co β†—2026-08-12verified
MMLU-Prohuggingface.co β†—2026-08-12verified
GPQA Diamondhuggingface.co β†—2026-08-12verified
SWE-bench Verifiedhuggingface.co β†—2026-08-12verified
Codeforces rating (Max)huggingface.co β†—2026-08-12verified
DeepSeek releases V4 preview β€” 1M context at open-weight frontier pricingapi-docs.deepseek.com β†—2026-05-25verified
How does V4-Flash compare to V4-Pro on real agentic workloads?codersera.com β†—2026-05-25to verify
Does V4-Flash support the same thinking mode API as V4-Pro?api-docs.deepseek.com β†—2026-05-25to verify
What changed in the July 2026 V4-Flash API update?api-docs.deepseek.com β†—2026-08-12to verify
Help & contact