Language

Qwen3.6-27B

Released
22 Apr 2026
Status
Legacy
Context
262.1K
Max output
65.5K
Input per 1M tokens
$0.60
Output per 1M tokens
$3.60
Open weights
Yes
License
apache-2.0
Parameters
27B
Knowledge cutoff
—
Reasoning
Yes
Modalities
text, image, video → text

Category scores

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

Category scores
CategoryScoreRankCoverage
Knowledge & reasoning29.3602 of 4

Benchmark results

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

Knowledge & reasoning

Knowledge & reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Chess PuzzlesPuzzles solved22.0%Independent—7 Aug 2026Epoch AI
Chess PuzzlesPuzzles solved9.0%Independentnone7 Aug 2026Epoch AI
DTBench (Decision Theory Benchmark)Accuracy (capability questions)78.1%Third party——https://conceptualreasoning.ai/dtbench via Epoch AI
Humanity's Last Exam (HLE)Accuracy24.0%Self-reported——Model Card via Hugging Face Community Evals
LMCA (Language Model Conceptual Argumentation)Correlation with expert ratings (Pearson r)0.3453Third party——https://conceptualreasoning.ai/lmca via Epoch AI
MMLU-ProAccuracy86.2%Self-reported——Model Card via Hugging Face Community Evals
Mystery Game PuzzlesPuzzles solved7.0%Independentnone27 Aug 2026Epoch AI

Scientific reasoning

Scientific reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GPQA DiamondAccuracy85.9%Independent—7 Aug 2026Epoch AI
GPQA DiamondAccuracy84.8%Independentnone7 Aug 2026Epoch AI
GPQA DiamondAccuracy87.8%Self-reported——Model Card via Hugging Face Community Evals

Mathematics

Mathematics results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
FrontierMath Tiers 1-3Accuracy35.1%Independent—28 Aug 2026Epoch AI
FrontierMath Tiers 1-3Accuracy34.0%Independentnone29 Aug 2026Epoch AI
OTIS Mock AIME 2024-2025Accuracy91.1%Independent—7 Aug 2026Epoch AI
OTIS Mock AIME 2024-2025Accuracy66.7%Independentnone7 Aug 2026Epoch AI

Other Qwen versions

Sources