RLTasksby ulam.ai
Models / Qwen

Qwen / MODEL EVIDENCE

Qwen3-235B-A22B

17 benchmarks · 18 observations · 17 sources · released 2025-04-28

Scores at a glance

17 benchmarks · large dot: Qwen3-235B-A22B; small dots: closest models where reported

GPQA Diamond ↗70.71%
9.2896.4

Qwen3-235B-A22B · GPQA diamond · Best score (across scorers) · % ↗

MATH Level 5 ↗68.86%
3.2998.13

Qwen3-235B-A22B · MATH level 5 · Best score (across scorers) · % ↗

LMCA ↗25%
2.7868.26

Qwen3-235B-A22B · LMCA · Score · % ↗

DTBench ↗75.73%
41.6598.93

Qwen3-235B-A22B · DTBench · Accuracy · % ↗

Fortress ↗61.39%
5.7874.39

Qwen3-235B-A22B · Scale Labs · Risk elicitation score ↗

FrontierMath-2025-02-28-Public ↗0%
070

qwen3-235b-a22b · FrontierMath-2025-02-28-Public · FrontierMath public · 10 questions · source run score · Epoch AI · public February 2025 sample · historical release ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

8.1 percentile-point gap

5 of 5 shared benchmark comparisons

9.6 percentile-point gap

4 of 5 shared benchmark comparisons

10.8 percentile-point gap

3 of 5 shared benchmark comparisons

How similarity is measured

Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.

Filter comparisons to this model →

Sources

Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.

  1. Aider code editing benchmark ↗

    Paul Gauthier, Aider contributors

  2. SealQA: Raising the Bar for Reasoning in Search-Augmented Language Models — published result table ↗

    Pham, Thinh, Nguyen, Nguyen, Zunjare, Pratibha, Chen, Weiyuan, Tseng, Yu-Min, Vu, Tu · 2025

  3. DTBench — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  4. LMCA — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  5. GPQA Diamond ↗

    Epoch AI

  6. MATH Level 5 ↗

    Epoch AI

  7. SimpleBench ↗

    Epoch AI

  8. Vending Bench 2 — methodology and published results ↗

    Epoch AI · 2026

  9. WeirdML (v2) ↗

    Epoch AI

  10. AI benchmarks & capabilities ↗

    Epoch AI

  11. Fiction.liveBench — March 14, 2025 benchmark release ↗

    Fiction.live · 2025

  12. Writing benchmark — incorporating mandatory story elements ↗

    lechmazur

  13. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  14. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  15. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  16. BrowseComp-ZH: Benchmarking Web Browsing Ability of Large Language Models in Chinese — published result table ↗

    Peilin Zhou, Bruce Leon, Xiang Ying, Can Zhang, Yifan Shao, Qichen Ye et al. · 2025

    All 16 authors

    Peilin Zhou, Bruce Leon, Xiang Ying, Can Zhang, Yifan Shao, Qichen Ye, Dading Chong, Zhiling Jin, Chenxuan Xie, Meng Cao, Yuxin Gu, Sixin Hong, Jing Ren, Jian Chen, Chao Liu, Yining Hua

  17. A Rosetta Stone for AI Benchmarks — released Epoch evaluation runs ↗

    Epoch AI · 2025

Names used in source tables

Qwen3 235B A22B Thinking · Qwen3-235B-A22B · Qwen3-235B-A22B (Non-Thinking) · Qwen3-235B-A22B (Thinking) · qwen3-235b-a22b