DeepSeek

DeepSeek V4 Pro

Open MoE flagship with million-token context for coding and long agent runs

deepseek/deepseek-v4-pro

Context

1M

Output

384K

Released

2026-04-24

Knowledge cutoff

2025-05

Types

InputText
OutputText

Reasoning

Supported

Tool use

Supported

Structured output

Supported

Attachments

—

Open weights

Supported

Provider pricing

OpenRouter
InText
OutText
Price
$0.30 / $0.59
Context
1M
Status
Stable
LLM Gateway
InText
OutText
Price
$0.44 / $0.87
Context
1.1M
Status
Stable
TokenGo
InText
OutText
Price
$0.44 / $0.87
Context
1M
Status
Stable
ZenMux
InText
OutText
Price
$0.44 / $0.87
Context
1M
Status
Stable
OrcaRouter
InText
OutText
Price
$0.44 / $0.88
Context
1M
Status
Stable
CommandCode
InText
OutText
Price
$0.66 / $1.98
Context
1M
Status
Stable
Eden AI
InText
OutText
Price
$0.66 / $1.98
Context
1M
Status
Stable
Tempr Gateway
InText
OutText
Price
$0.66 / $1.98
Context
1M
Status
Stable
Vercel AI Gateway
InText
OutText
Price
$0.66 / $1.98
Context
1M
Status
Stable
NanoGPT
InText
OutText
Price
$1.10 / $2.20
Context
1M
Status
Stable
AI/ML API
InText
OutText
Price
$1.17 / $3.51
Context
1M
Status
Stable
Merge Gateway
InText
OutText
Price
$1.32 / $3.96
Context
1M
Status
Stable
Ofox
InText
OutText
Price
$1.32 / $3.96
Context
1M
Status
Stable
CrossModel
InText
OutText
Price
$1.35 / $4.05
Context
1M
Status
Stable
Kilo Gateway
InText
OutText
Price
$1.60 / $3.20
Context
1M
Status
Stable
Novita AI
InText
OutText
Price
$1.60 / $3.20
Context
1M
Status
Stable
Arcee
InText
OutText
Price
$1.74 / $3.48
Context
512K
Status
beta
Price
$1.74 / $3.48
Context
131.1K
Status
Stable
FastRouter
InText
OutText
Price
$1.74 / $3.48
Context
1M
Status
Stable
HPC-AI
InText
OutText
Price
$1.74 / $3.48
Context
1M
Status
Stable
Impossibl
InText
OutText
Price
$1.74 / $3.48
Context
1M
Status
Stable
TensorX
InText
OutText
Price
$1.75 / $3.50
Context
1M
Status
Stable
Prime Inference
InText
OutText
Price
$1.91 / $3.83
Context
1M
Status
Stable
AnyAPI
InText
OutText
Price
—
Context
1M
Status
Stable

Benchmarks

  • SWE-Bench Verified

    80.6 · resolved

  • Artificial Analysis Coding Agent Index

    50.1 · average pass@1

  • SWE-Atlas Codebase QnA

    67.8 · pass@1

  • SWE-Bench Pro

    18 · pass@1

  • Terminal-Bench

    64.7 · pass@1

  • MMLU-Pro

    87.5 · EM

  • SimpleQA-Verified

    57.9 · pass@1

  • Chinese SimpleQA

    84.4 · pass@1

  • GPQA Diamond

    90.1 · pass@1

  • Humanity's Last Exam

    37.7 · pass@1

  • LiveCodeBench

    93.5 · pass@1

  • Codeforces

    3206 · rating

  • HMMT

    95.2 · pass@1

  • IMOAnswerBench

    89.8 · pass@1

  • MathArena Apex

    38.3 · pass@1

  • MathArena Apex Shortlist

    90.2 · pass@1

  • MRCR

    83.5 · MMR

  • CorpusQA

    62 · accuracy

  • Terminal-Bench

    67.9 · accuracy

  • SWE-Bench Pro

    55.4 · resolved

  • SWE-Bench Multilingual

    76.2 · resolved

  • BrowseComp

    83.4 · pass@1

  • Humanity's Last Exam

    48.2 · pass@1

  • MCP Atlas

    73.6 · pass@1

  • GDPval-AA

    1554 · Elo

  • Toolathlon

    51.8 · pass@1

From the same lab