supermarket

Language

Qwen3.5-9B

By Alibaba Group, in the Qwen family.

Released
2 Mar 2026
Status
Legacy
Context
262.1K
Max output
—
Input per 1M tokens
—
Output per 1M tokens
—
Open weights
Yes
License
apache-2.0
Parameters
9B
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 & reasoning18.1812 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 solved12.0%Independentnone27 Aug 2026Epoch AI
Chess PuzzlesPuzzles solved2.0%Independent—27 Aug 2026Epoch AI
DTBench (Decision Theory Benchmark)Accuracy (capability questions)71.2%Third party——https://conceptualreasoning.ai/dtbench via Epoch AI
LMCA (Language Model Conceptual Argumentation)Correlation with expert ratings (Pearson r)0.2447Third party——https://conceptualreasoning.ai/lmca via Epoch AI

Scientific reasoning

Scientific reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GPQA DiamondAccuracy79.0%Independent—27 Aug 2026Epoch AI
GPQA DiamondAccuracy78.9%Independentnone27 Aug 2026Epoch AI

Mathematics

Mathematics results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
OTIS Mock AIME 2024-2025Accuracy61.7%Independentnone27 Aug 2026Epoch AI
OTIS Mock AIME 2024-2025Accuracy42.8%Independent—27 Aug 2026Epoch AI

Coding & software engineering

Coding & software engineering results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Terminal-Bench 2.0Accuracy (mean over trials)9.2%Third party——https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI

Other Qwen versions

Sources