Qwen3-235B-A22B-Instruct-2507: benchmark scores, price and ranking
Qwen3-235B-A22B-Instruct-2507 from Alibaba ranks #278 of 743 LLMs on the LLMs Tiger aggregate benchmark score (49.3), based on 4 benchmark results captured up to 2026-10-13.
- Organisation
- Alibaba
- Type
- Open-weight / local model
- Open weights
- Yes
- Released
- 2025-07-25 (source: epoch)
- Cheapest API price
- $0.090 in / $0.550 out per million tokens (via deepinfra, observed 2026-10-13T20:47Z)
- Aggregate rank
- #278 of 743 (score 49.3)
Relative to other models, Qwen3-235B-A22B-Instruct-2507 is strongest on MMLU-Pro (#22 of 136) and weakest on LMCA (#67 of 108).
Across the 4 benchmarks where at least ten models are scored, Qwen3-235B-A22B-Instruct-2507 is in the top 10% on 0 and in the top half on 2.
Its cheapest listed input price, $0.090 per million tokens, is lower than that of 83% of the 427 models with a listed price.
Qwen3-235B-A22B-Instruct-2507 benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| MMLU-Pro | 83.0% | #22 of 136 | HuggingFace MMLU-Pro leaderboard | 2026-10-13 |
| DTBench | 78.4% | #58 of 146 | Epoch AI Benchmarking Hub | 2026-10-13 |
| WeirdML | 38.7% | #67 of 109 | Epoch AI Benchmarking Hub | 2026-10-13 |
| LMCA | 23.5% | #67 of 108 | Epoch AI Benchmarking Hub | 2026-10-13 |
About Qwen3-235B-A22B-Instruct-2507
Qwen3-235B-A22B-Instruct-2507 is an open-weight model that can run on your own hardware from Alibaba. It was released on 2025-07-25, according to Epoch AI. The cheapest API listing we track charges $0.090 per million input tokens and $0.550 per million output tokens, through deepinfra. We hold 4 benchmark results for it across 3 areas, and none is in the top tenth of its board.
Where Qwen3-235B-A22B-Instruct-2507 stands
On our aggregate score, Qwen3-235B-A22B-Instruct-2507 is #278 of 743 ranked models, around the middle of the table. Its score is 49.3; the scale is centred on 50, so it sits 0.7 points below the typical model. That beats 63% of the ranked field. Among the 74 ranked Alibaba models it is #26. The next one up is Qwen3.5 27B at 49.3. The next one down is qwq-plus at 49.3.
Qwen3-235B-A22B-Instruct-2507, benchmark by benchmark
MMLU-Pro puts Qwen3-235B-A22B-Instruct-2507 at 83.0%, #22 of 136. That beats 84% of the models tested, but 5.1 points short of Darwin-180B-RSI (88.1%). The results either side are EXAONE-4.5-33B (83.3%) and LongCat-Flash-Chat (82.7%). It covers 57-subject knowledge + reasoning. The board median is 55.3%, so Qwen3-235B-A22B-Instruct-2507 is 27.7 points above it. Source: HuggingFace MMLU-Pro leaderboard, captured 2026-10-13.
On DTBench, Qwen3-235B-A22B-Instruct-2507 reaches 78.4%. It ranks #58 of 146, ahead of 60% of the field, and 20.5 points behind Claude Opus 5.5 (max) at 98.9%. The results either side are MiniMax-M3 (78.9%) and Qwen3.6 27B (78.1%). It covers dynamic tool-use benchmark. The board median is 68.8%, so Qwen3-235B-A22B-Instruct-2507 is 9.6 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Qwen3-235B-A22B-Instruct-2507 lands below the middle on WeirdML: 38.7%, #67 of 109, with 61% of the field scoring higher. GPT-6 Astra (max) leads at 93.3%; the closest result above is GPT-4.1 at 39.0%. It covers unusual machine-learning tasks. The board median is 41.2%, so Qwen3-235B-A22B-Instruct-2507 is 2.5 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Qwen3-235B-A22B-Instruct-2507 lands below the middle on LMCA: 23.5%, #67 of 108, with 61% of the field scoring higher. Claude Opus 5.5 (max) leads at 68.2%; the closest result above is qwen-plus-2025-04-28 at 24.0%. It covers language-model capability assessment. The board median is 29.2%, so Qwen3-235B-A22B-Instruct-2507 is 5.7 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
General knowledge and chat: uneven, ahead of 38% of the field on average across 2 benchmarks. Its best board here is WeirdML (#67 of 109) and its weakest is LMCA (#67 of 108).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so Qwen3-235B-A22B-Instruct-2507 costs less than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), DeepSeek V4 Flash 0731 (max) ($0.016, 53.1). Similar scores cost more on Reka Flash 3 ($0.100) and Qwen3.8 Flash (xhigh) ($0.113).
Release history
The Alibaba model released before it was Qwen3-Coder-480B-A35B-Instruct on 2025-07-22, which scores 45.5. The next one was qwen3-30b-a3b-instruct-2507 on 2025-07-29, at 46.8.
Reading these numbers
The 4 scores on this page come from 2 sources: Epoch AI Benchmarking Hub (3), HuggingFace MMLU-Pro leaderboard (1). No official leaderboard supplied any of them, so treat them as third-party measurements. With 4 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #275 gemini-2.5-flash-preview-09-2025_16k (49.3)
- #276 Qwen3-4B-Thinking-2507 (49.3)
- #277 Qwen3.5 27B (49.3)
- #279 qwq-plus (49.3)
- #280 Mistral-Large-Instruct-2407 (49.3)
- #281 gemini-2.5-flash-preview-05-20_16K (49.3)
More from Alibaba
- Qwen 3.6 Max (Preview) (#25, 54.1)
- Qwen3.8 Max (xhigh) (#37, 53.2)
- Qwen3.7 Max (#71, 52.3)
- ollama/qwen2.5-coder:32b (#93, 51.9)
- Qwen3 235B A22B diff, no think, Alibaba API (#122, 51.5)
- Qwen2.5-Coder-14B-Instruct (#135, 51.3)
Compare it on the live leaderboard → · How the aggregate score works