Qwen3 235B A22B 2507 (Reasoning)
64±0
64interval65
aiqbench · LLM benchmarks
Benchmarks
- Sources
- 8
- Entries tracked
- 22,310
- Last sync
- 1 min1 minute ago
All sources syncing normally
- Arena · Text· 2 hours ago
- Arena · WebDev· 2 hours ago
- Arena · Vision· 2 hours ago
- Arena · Agent· 2 hours ago
- GPQA Diamond· 1 minute ago
- SWE-bench· 1 minute ago
- Terminal-Bench· 1 minute ago
- ARC-AGI-2· 1 minute ago
Filter by org:
Released:
Qwen3 235B A22B 2507 Instruct
57±0
57interval58
qwen3-235b-a22b-thinking-2507
55±2
53interval58
One grid, every column. Pick axes in the Axes panel — one axis ranks like a single board, several blend into a weighted composite, all of them read as the index. Rank is assigned before any filter and never renumbers.
| Model | Org | Interval | Swing | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | Add to compare | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 80 | Qwen3 235B A22B 2507 (Reasoning) | alibaba | Jul 2025 | $0.09 / $0.55 | 64.4 78.8Ranked 64.4 = (3 × 78.8 + 3 × 50) ÷ 6. 78.8 is the mean percentile over the 3 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50. | — | 30 | 3/22 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 89 | 89 | 59 | — | ||
| 211 | Qwen3 235B A22B 2507 Instruct | alibaba | — | — | 57.3 64.7Ranked 57.3 = (3 × 64.7 + 3 × 50) ÷ 6. 64.7 is the mean percentile over the 3 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50. | — | 30 | 3/22 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 82 | 59 | 52 | — | ||
| 270 | qwen3-235b-a22b-thinking-2507 | alibaba | Jul 2025 | $0.45 / $3.50 | 55.2 58.3Ranked 55.2 = (5 × 58.3 + 3 × 50) ÷ 8. 58.3 is the mean percentile over the 5 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50. | 53–58 | 24 | 5/22 | #118–270 | 62 | — | — | — | — | — | — | 61 | 67 | — | 43 | — | — | — | 58 | — | — | — | — | — | — | — | ||
| 288 | qwen3-235b-a22b-thinking | Alibaba | — | — | 54.7 68.7Ranked 54.7 = (1 × 68.7 + 3 × 50) ÷ 4. 68.7 is the mean percentile over the 1 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50. | — | — | 1/22low | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 69 | — | — | — | — | — | — | — | ||
| 320 | qwen3-235b-a22b-no-thinking | alibaba | Oct 2026 | — | 53.3 63.4Ranked 53.3 = (1 × 63.4 + 3 × 50) ÷ 4. 63.4 is the mean percentile over the 1 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50. | 52–54 | — | 1/22low | #114–320 | 63 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | ||
| 343 | qwen3-235b-a22b | alibaba | Apr 2025 | — | 52.4 53.5Ranked 52.4 = (7 × 53.5 + 3 × 50) ÷ 10. 53.5 is the mean percentile over the 7 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50. | 52–53 | 57 | 7/22 | #136–343 | 56 | — | — | — | — | — | — | 52 | — | — | — | — | — | 25 | — | — | — | — | 82 | 66 | 37 | 56 | ||
| 384 | qwen3-235b-a22b-instruct-2507 | alibaba | Jul 2025 | — | 51.1 51.8Ranked 51.1 = (5 × 51.8 + 3 × 50) ÷ 8. 51.8 is the mean percentile over the 5 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50. | 50–52 | 57 | 5/22 | #82–384 | 71 | — | — | 40 | — | — | — | — | — | — | — | 20 | — | — | 51 | 77 | — | — | — | — | — | — | ||
| 478 | Qwen/Qwen3-235B-A22B-Thinking-2507 | Alibaba | — | — | 48.4 43.5Ranked 48.4 = (1 × 43.5 + 3 × 50) ÷ 4. 43.5 is the mean percentile over the 1 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50. | 47–49 | — | 1/22low | — | — | — | — | — | — | — | — | — | — | 43 | — | — | — | — | — | — | — | — | — | — | — | — |
Blend is coverage-shrunk; the raw composite is shown greyed beside it · A low badge means fewer than 3 boards scored this model