qwen3-30b-a3b-instruct-2507: benchmark scores, price and ranking
qwen3-30b-a3b-instruct-2507 from Alibaba ranks #482 of 743 LLMs on the LLMs Tiger aggregate benchmark score (46.8), based on 5 benchmark results captured up to 2026-10-13.
- Organisation
- Alibaba
- Type
- Open-weight / local model
- Open weights
- Yes
- Released
- 2025-07-29 (source: epoch)
- Cheapest API price
- $0.100 in / $0.300 out per million tokens (via nebius, observed 2026-10-13T20:47Z)
- Aggregate rank
- #482 of 743 (score 46.8)
Relative to other models, qwen3-30b-a3b-instruct-2507 is strongest on OTIS Mock AIME (#91 of 189) and weakest on Chess Puzzles (#91 of 122).
Across the 5 benchmarks where at least ten models are scored, qwen3-30b-a3b-instruct-2507 is in the top 10% on 0 and in the top half on 1.
Its cheapest listed input price, $0.100 per million tokens, is lower than that of 79% of the 427 models with a listed price.
qwen3-30b-a3b-instruct-2507 benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| OTIS Mock AIME | 62.2% | #91 of 189 | Epoch AI Benchmarking Hub | 2026-10-13 |
| DTBench | 67.2% | #81 of 146 | Epoch AI Benchmarking Hub | 2026-10-13 |
| LMCA | 22.4% | #68 of 108 | Epoch AI Benchmarking Hub | 2026-10-13 |
| GPQA Diamond | 55.6% | #177 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Chess Puzzles | 2.0% | #91 of 122 | Epoch AI Benchmarking Hub | 2026-10-13 |
About qwen3-30b-a3b-instruct-2507
qwen3-30b-a3b-instruct-2507 is an open-weight model that can run on your own hardware from Alibaba. It was released on 2025-07-29, according to Epoch AI. The cheapest API listing we track charges $0.100 per million input tokens and $0.300 per million output tokens, through nebius. We hold 5 benchmark results for it across 3 areas, and none is in the top tenth of its board.
Where qwen3-30b-a3b-instruct-2507 stands
On our aggregate score, qwen3-30b-a3b-instruct-2507 is #482 of 743 ranked models, in the lower half. Its score is 46.8; the scale is centred on 50, so it sits 3.2 points below the typical model. That beats 35% of the ranked field. Among the 74 ranked Alibaba models it is #48. The next one up is Qwen3-32B at 46.9. The next one down is qwen3.6-plus at 46.6.
qwen3-30b-a3b-instruct-2507, benchmark by benchmark
OTIS Mock AIME puts qwen3-30b-a3b-instruct-2507 at 62.2%, #91 of 189. That beats 52% of the models tested, but 37.8 points short of Claude Fable 5.1 (max) (100.0%). The results either side are Qwen3-30B-A3B (62.8%) and Qwen3.5 9B (61.7%). It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so qwen3-30b-a3b-instruct-2507 is 4.4 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
qwen3-30b-a3b-instruct-2507 lands below the middle on DTBench: 67.2%, #81 of 146, with 55% of the field scoring higher. Claude Opus 5.5 (max) leads at 98.9%; the closest result above is Qwen3-32B at 67.5%. It covers dynamic tool-use benchmark. The board median is 68.8%, so qwen3-30b-a3b-instruct-2507 is 1.6 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
qwen3-30b-a3b-instruct-2507 lands below the middle on LMCA: 22.4%, #68 of 108, with 62% of the field scoring higher. Claude Opus 5.5 (max) leads at 68.2%; the closest result above is Qwen3-235B-A22B-Instruct-2507 at 23.5%. It covers language-model capability assessment. The board median is 29.2%, so qwen3-30b-a3b-instruct-2507 is 6.9 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
qwen3-30b-a3b-instruct-2507 lands below the middle on GPQA Diamond: 55.6%, #177 of 275, with 64% of the field scoring higher. GPT-6 Astra (max) leads at 95.8%; the closest result above is DeepSeek-R1-Distill-Llama-70B at 55.7%. It covers expert-level science questions. The board median is 71.7%, so qwen3-30b-a3b-instruct-2507 is 16.1 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
qwen3-30b-a3b-instruct-2507 lands below the middle on Chess Puzzles: 2.0%, #91 of 122, with 74% of the field scoring higher. GPT-6 Astra (max) leads at 72.0%; the closest result above is Magistral Small 1.2 at 3.0%. It covers chess tactics puzzles. The board median is 12.5%, so qwen3-30b-a3b-instruct-2507 is 10.5 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
Reasoning, maths and science: uneven, ahead of 44% of the field on average across 2 benchmarks. Its best board here is OTIS Mock AIME (#91 of 189) and its weakest is GPQA Diamond (#177 of 275). General knowledge and chat: uneven, ahead of 31% of the field on average across 2 benchmarks. Its best board here is LMCA (#68 of 108) and its weakest is Chess Puzzles (#91 of 122).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so qwen3-30b-a3b-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), Qwen2.5-Coder-3B-Instruct ($0.010, 46.5). Similar scores cost more on MiMo-V2-Flash ($0.110) and qwen3-coder-next ($0.120).
Release history
The Alibaba model released before it was Qwen3-235B-A22B-Instruct-2507 on 2025-07-25, which scores 49.3. The next one was parasail-qwen3-235b-a22b-instruct-2507 on 2025-09-01, at 49.0.
Reading these numbers
The 5 scores on this page come from 1 source: Epoch AI Benchmarking Hub (5). No official leaderboard supplied any of them, so treat them as third-party measurements. With 5 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #479 GPT-4 (1106) (46.9)
- #480 glm-4-9b (46.9)
- #481 Llama-4-Maverick-17B-128E-Instruct (46.9)
- #483 Command A+ (46.8)
- #484 MiniMax-M2.5 (46.8)
- #485 o4-mini (high) (46.8)
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