supermarket

Language

DeepSeek-V4.1-Flash

By DeepSeek, in the DeepSeek-V family.

Released
10 Sep 2026
Status
Current
Context
1M
Max output
393.2K
Input per 1M tokens
$0.30
Output per 1M tokens
$1.20
Open weights
Yes
License
mit
Parameters
552B
Knowledge cutoff
—
Reasoning
Yes
Modalities
text, image → text

Category scores

Out of 100, from independent and third-party results only, by method category-2026-09-29.1.

Category scores
CategoryScoreRankCoverage
Knowledge & reasoning55.3Not ranked1 of 4
Human preference69.5Not ranked2 of 5

Benchmark results

Every result we hold for this model, with who ran it and where it was published.

Knowledge & reasoning

Knowledge & reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
DTBench (Decision Theory Benchmark)Accuracy (capability questions)89.9%Third partyunknown—https://conceptualreasoning.ai/dtbench via Epoch AI
Humanity's Last Exam (HLE)Accuracy63.9%Self-reported——DeepSeek-V4.1-Flash model card via Hugging Face Community Evals
LMCA (Language Model Conceptual Argumentation)Correlation with expert ratings (Pearson r)0.4697Third partyunknown—https://conceptualreasoning.ai/lmca via Epoch AI

Scientific reasoning

Scientific reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GPQA DiamondAccuracy90.9%Self-reported——DeepSeek-V4.1-Flash model card via Hugging Face Community Evals
Surface Evolver BenchMean score (partial credit)46.3%Third partyhigh—https://yhenon.github.io/surface-evolver-llm-eval/ via Epoch AI

Human preference

Human preference results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Code Arena: WebDev (arena.ai)Arena score (Elo scale)1621Independentmax25 Sep 2026Arena
Code Arena: WebDev (arena.ai)Votes3815Independentmax25 Sep 2026Arena
Text Arena (arena.ai)Arena score (Elo scale)1477Independentmax25 Sep 2026Arena
Text Arena (arena.ai)Votes6630Independentmax25 Sep 2026Arena

Other DeepSeek-V versions

Sources

DeepSeek-V4.1-Flash: benchmarks, price and specs · supermarket.dev