Model record

DeepSeek V4 Flash 0731

maxOpen weights

DeepSeek · released Jul 31, 2026

Overall standing

Overall
55.4rank 17 of 63
55.4 is at approximately the 74th percentile of the ranked Overall field. The field spans 19.0 to 62.2, with a median of 49.4.
7 of 8 measured · 1 estimated
Category standingsReasoning 66.7Coding 64.3Math 16.6Agentic 31.1
Reasoning
66.7rank 25 of 63
66.7 is at approximately the 61th percentile of the ranked Reasoning field. The field spans 31.6 to 73.3, with a median of 65.0.
3 of 4 measured · 1 estimated
Coding
64.3rank 20 of 63
64.3 is at approximately the 69th percentile of the ranked Coding field. The field spans 10.4 to 72.4, with a median of 54.5.
2 of 2 measured
Math
16.6rank 14 of 63
16.6 is at approximately the 79th percentile of the ranked Math field. The field spans 0.0 to 32.3, with a median of 5.7.
1 of 1 measured
Agentic
31.1rank 6 of 63
31.1 is at approximately the 91th percentile of the ranked Agentic field. The field spans 3.3 to 33.4, with a median of 19.0.
1 of 1 measured
Model factsOpen weights · $0.14 / $0.28 per 1M

Model facts come from the provider page. Listed API prices carry a separate first-party source and check date.

Provider
DeepSeek
Released
Weights
Open weights
Context window
1,000,000 tokens
Input price / Mtok
$0.14
Output price / Mtok
$0.28
Price source
Official pricing (opens in a new tab)
Reasoning effort
reasoning, max effort

Scores and sources

Measured scores retain their source and retrieval date. Missing scores are not treated as zero.

Benchmark scores, evaluation settings, sources, and retrieval dates for DeepSeek V4 Flash 0731
BenchmarkCategoryScoreEvaluation settingsSource
Terminal-Bench v2.1Coding78.7Artificial Analysis independent run; 89 tasks; Terminus 2 on E2B; 3 repeats; pass@1; 250-episode cap; 2-hour timeoutOpen source (opens in a new tab)
τ³-BankingAgentic31.1Artificial Analysis independent run; 97 tasks; 5 repeats; backend-state pass@1; BM25 plus grep retrieval; 200-step capOpen source (opens in a new tab)
AA-LCRReasoning65.7Artificial Analysis independent run; 100 roughly 100k-token multi-document questions; 3 repeats; equality-checker pass@1; no toolsOpen source (opens in a new tab)
Humanity's Last ExamReasoning36.8Artificial Analysis independent run; 2,158 text-only questions; pass@1; GPT-4o (August 2024) equality checker; no toolsOpen source (opens in a new tab)
GPQA DiamondReasoning90.8Artificial Analysis independent run; 198 questions; 5 repeats; regex-graded pass@1; no toolsOpen source (opens in a new tab)
SciCodeCoding49.9Artificial Analysis independent run; 288 test subproblems; scientist background included; 3 repeats; subproblem pass@1Open source (opens in a new tab)
IFBenchReasoning—Not measured in this datasetNo source
CritPtMath16.6Artificial Analysis independent run; 70 test challenges; 5 repeats; two-step answer parsing; official CritPt grader; pass@1; no toolsOpen source (opens in a new tab)