Recent studyBlind TTS Elo is live. Compare two anonymous voice samples, vote after listening, and help separate real preference signal from noise.Vote in the study ->
Codesota · Models · Qwen3.6-27B11 results · 11 benchmarks
Model card

Qwen3.6-27B.

unknown
§ 02 · Benchmarks

Every benchmark Qwen3.6-27B has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01Video-MMEMultimodal · Video Understandingaccuracy87.7%#1/24source ↗
02MMMUMultimodal · Image-Text-to-Textaccuracy82.9%#3/36source ↗
03RealWorldQAMultimodal · Visual Question Answeringaccuracy84.1%#3/23source ↗
04MMStarMultimodal · Image-Text-to-Textaccuracy81.4%#4/21source ↗
05MVBenchMultimodal · Video Understandingaccuracy75.5%#5/20source ↗
06LiveCodeBenchComputer Code · Code Generationpass-183.9%#6/24source ↗
07SWE-Bench VerifiedComputer Code · Code Generationaccuracy77.2%#7/22source ↗
08MMMU-ProMultimodal · Visual Question Answeringaccuracy75.8%#10/31source ↗
09GPQA DiamondReasoning · Multi-step Reasoningaccuracy87.8%#11/74source ↗
10MMLU-ProReasoning · Commonsense Reasoningaccuracy86.2%#11/73source ↗
11HLEReasoning · Multi-step Reasoningaccuracy24.0%#17/36source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area

Where Qwen3.6-27B actually performs.

Multimodal
6
benchmarks
avg rank #4.3
Computer Code
2
benchmarks
avg rank #6.5
Reasoning
3
benchmarks
avg rank #13.0
§ 04 · Papers

1 paper with results for Qwen3.6-27B.

  1. 2026-04-21· 11 results

    Qwen3.6

§ 06 · Sources & freshness

Where these numbers come from.

pwc-dump
11
results
0 of 11 rows marked verified.