Qwen3 235B-A22B Thinking 2507
| Released | 2025-07-25 |
|---|---|
| Price per 1M | $0.30 in · $2.90 out |
| Overall score | 44.3 · rank 164 |
| Context window | 262,144 tokens |
| Provider | Alibaba |
| Weights | Open weights (Apache 2.0) |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 44.3 | 164 | 10 |
| Overall: Coding | 43.4 | 169 | 5 |
| Overall: Reasoning | 48.1 | 131 | 7 |
| Overall: Agentic work | 37.1 | 205 | 5 |
| Overall: Frontend | 47.5 | 116 | 1 |
| Coding | 47.5 | 136 | 1 |
| Agents | 30.6 | 182 | 1 |
| Reasoning | 45.8 | 147 | 3 |
| Math | 48.0 | 129 | 3 |
| Knowledge | 53.4 | 74 | 1 |
| General | 48.5 | 124 | 1 |
| Value vs price | −3.3 | 150 |
Evidence
Coding 47.5 ±4.7
Agents 30.6 ±7.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Vending-Bench 2 | $-11 | Epoch AI | 2026-11-11 |
Reasoning 45.8 ±6.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| GPQA Diamond | 80.1% | Epoch AI | 2025-12-11 |
| CritPt | 0.0% | Epoch AI | 2026-11-11 |
| Mystery Game Puzzles | 9.0% | Epoch AI | 2026-08-27 |
Math 48.0 ±4.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierMath Tiers 1-3 (2025 set) | 8.5% | Epoch AI | 2025-12-11 |
| FrontierMath Tier 4 (2025 set) | 0.0% | Epoch AI | 2025-12-11 |
| OTIS Mock AIME 2024-2025 | 86.7% | Epoch AI | 2025-12-10 |
Knowledge 53.4 ±3.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| SimpleQA Verified | 40.4% | Epoch AI | 2026-08-27 |
General 48.5 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Text | 1,400 | LMArena | 2026-11-09 |
Names matched to this model
fireworks-models-qwen3-235b-a22b-thinking-2507, qwen3-235b-a22b-thinking-2507