Qwen3-235B-A22B-Thinking (Jul 2025) benchmark ranking on every AI leaderboard
open weights · top 10 on 0 of 7 · released Jul 25, 2025
| Rank | Board | Score | |||
|---|---|---|---|---|---|
| 20 |  BFCLAgents and tools · as “Qwen3-235B-A22B-Instruct-2507 (Prompt)” | Agents and tools | 52.1% | 81 | |
| 27 |  MultiChallengeChat · as “qwen3-235b-a22b” | Chat | 41.2% | 29 | |
| 63 |  ARC-AGI-2Reasoning · as “Qwen3-235b-a22b Instruct (25/07)” | Reasoning | 1.3% | 78 | |
| 72 |  GPQA DiamondReasoning · as “Qwen3-235B-A22B-Thinking-2507” | Reasoning | 80.1% | 100 | |
| 85 | Epoch ECIReasoning · as “Qwen3-235B-A22B-Thinking (Jul 2025)” | Reasoning | 143.9 | 100 | |
| 92 |  Text ArenaChat · as “qwen3-235b-a22b-instruct-2507” | Chat | 1,423 | 100 | |
| 37 |  MMLU-ProReasoning · stale since 2026-03-11 · as “Qwen3-235B-A22B-Thinking-2507” | Reasoning | 84.5% | 100 |
Category ranks and more models
Scores as published by each board on the capture date. Model names and logos belong to their owners; logos via logo.dev.