qwen2.5-32b-instruct: benchmark scores, price and ranking
qwen2.5-32b-instruct from Alibaba ranks #611 of 743 LLMs on the LLMs Tiger aggregate benchmark score (45.0), based on 3 benchmark results captured up to 2026-10-13.
- Organisation
- Alibaba
- Type
- Open-weight / local model
- Open weights
- Yes
- Released
- 2024-09-17 (source: epoch)
- Cheapest API price
- $0.060 in / $0.200 out per million tokens (via nebius, observed 2026-10-13T20:47Z)
- Aggregate rank
- #611 of 743 (score 45.0)
Relative to other models, qwen2.5-32b-instruct is strongest on GPQA Diamond (#204 of 275) and weakest on Chess Puzzles (#120 of 122).
Across the 3 benchmarks where at least ten models are scored, qwen2.5-32b-instruct is in the top 10% on 0 and in the top half on 0.
Its cheapest listed input price, $0.060 per million tokens, is lower than that of 89% of the 427 models with a listed price.
qwen2.5-32b-instruct benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| GPQA Diamond | 46.1% | #204 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| OTIS Mock AIME | 7.4% | #145 of 189 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Chess Puzzles | 0.0% | #120 of 122 | Epoch AI Benchmarking Hub | 2026-10-13 |
About qwen2.5-32b-instruct
qwen2.5-32b-instruct is an open-weight model that can run on your own hardware from Alibaba. It was released on 2024-09-17, according to Epoch AI. The cheapest API listing we track charges $0.060 per million input tokens and $0.200 per million output tokens, through nebius. We hold 3 benchmark results for it across 2 areas, and none is in the top tenth of its board.
Where qwen2.5-32b-instruct stands
On our aggregate score, qwen2.5-32b-instruct is #611 of 743 ranked models, in the lower half. Its score is 45.0; the scale is centred on 50, so it sits 5.0 points below the typical model. That beats 18% of the ranked field. Among the 74 ranked Alibaba models it is #65. The next one up is Qwen2.5-Coder-1.5B-Instruct at 45.4. The next one down is Qwen3-30B-A3B at 44.9.
qwen2.5-32b-instruct, benchmark by benchmark
qwen2.5-32b-instruct lands below the middle on GPQA Diamond: 46.1%, #204 of 275, with 74% of the field scoring higher. GPT-6 Astra (max) leads at 95.8%; the closest result above is Llama-3.1-Tulu-3-70B-DPO at 46.3%. It covers expert-level science questions. The board median is 71.7%, so qwen2.5-32b-instruct is 25.6 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
qwen2.5-32b-instruct lands below the middle on OTIS Mock AIME: 7.4%, #145 of 189, with 76% of the field scoring higher. Claude Fable 5.1 (max) leads at 100.0%; the closest result above is gemma-3-4b-it at 7.5%. It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so qwen2.5-32b-instruct is 50.4 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Chess Puzzles is a weak spot. qwen2.5-32b-instruct scores 0.0%, #120 of 122, behind 98% of the models measured. The leader, GPT-6 Astra (max), reaches 72.0%. It covers chess tactics puzzles. The board median is 12.5%, so qwen2.5-32b-instruct is 12.5 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
Reasoning, maths and science: weak, ahead of 25% of the field on average across 2 benchmarks. Its best board here is GPQA Diamond (#204 of 275) and its weakest is OTIS Mock AIME (#145 of 189).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so qwen2.5-32b-instruct 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), Qwen2.5-Coder-3B-Instruct ($0.010, 46.5). Similar scores cost more on Qwen3.5 Flash ($0.065) and phi-4 ($0.070).
Release history
The Alibaba model released before it was Qwen2.5-32B on 2024-09-17, which scores 49.7. The next one was ollama/qwen2.5-coder:14b on 2024-09-18, at 50.1.
Reading these numbers
The 3 scores on this page come from 1 source: Epoch AI Benchmarking Hub (3). No official leaderboard supplied any of them, so treat them as third-party measurements. With 3 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #608 DeepSeek-Coder-V2-Lite-Base (45.0)
- #609 mistral-small-3-2501 (45.0)
- #610 GLM-4.6 (45.0)
- #612 Qwen3-30B-A3B (44.9)
- #613 aya-expanse-8b (44.9)
- #614 Mercury 2 (44.9)
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