qwen3-30b-a3b-thinking-2507
70.1%±2.8%
67.2%CI72.9%
aiqbench · LLM benchmarks
Benchmarks
- Sources
- 8
- Entries tracked
- 22,310
- Last sync
- 31 min31 minutes ago
All sources syncing normally
- Arena · Text· 5 hours ago
- Arena · WebDev· 5 hours ago
- Arena · Vision· 5 hours ago
- Arena · Agent· 5 hours ago
- GPQA Diamond· 31 minutes ago
- SWE-bench· 31 minutes ago
- Terminal-Bench· 31 minutes ago
- ARC-AGI-2· 31 minutes ago
Filter by org:
Released:
Qwen3-30B-A3B
61.7%±3.0%
58.7%CI64.7%
qwen3-30b-a3b-instruct-2507
55.6%±2.9%
52.7%CI58.5%
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 | ± CI | Swing | Add to compare | |||||
|---|---|---|---|---|---|---|---|---|---|
| 108 | qwen3-30b-a3b-thinking-2507 | Alibaba | Jul 2025 | $0.20 / $2.40 | 70.1% | 67.2%–72.9% | #210–400 | ||
| 126 | Qwen3-30B-A3B | Alibaba | Apr 2025 | $0.12 / $0.50 | 61.7% | 58.7%–64.7% | #187–453 | ||
| 138 | qwen3-30b-a3b-instruct-2507 | Alibaba | Jul 2025 | $0.09 / $0.30 | 55.6% | 52.7%–58.5% | #123–410 |
Showing GPQA Diamond exactly as the source published it — its own score and confidence interval, no blending. Select a second axis to build a weighted composite.