AlibabaQwen3 Vl 32B
Best configuration ranks #333 of 610 on the BenchLeader Index at 48.2 ±5.7 (thinking reasoning effort). Released 21 Oct 2025.
- Blended price
- $0.838/M
- $0.160 in · $2.87 out
- Output speed
- 90 tok/s
- First answer
- 25 s
- first token 2.68 s
- Context
- 131k
- Overall index48
- Agents & tools40
- Knowledge39
- Instruction following59
- Multimodal48
- Long context54
- Composite44
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Default” means the publisher did not say which setting was used.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| thinkingbest | 48.2 | #333 | 90 tok/s | 25 s | $0.0009 | Agents & tools 40 · Composite 44 · Instruction following 59 · Knowledge 39 · Long context 54 · Multimodal 48 |
| default | 44.3 | #420 | 66 tok/s | 2.68 s | $0.0003 | Agents & tools 41 · Composite 40 · Instruction following 42 · Knowledge 34 · Multimodal 49 |
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | thinking | default | Source |
|---|---|---|---|
| GPQA Diamond (AA)not in index | 73.3%#258 | 67.1%#309 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 10.2%#255 | 6.8%#308 | Artificial Analysis |
Agents & tools
| Benchmark | thinking | default | Source |
|---|---|---|---|
| Terminal-Bench Hard | 7.6%#232 | 8.3%#226 | Artificial Analysis |
| τ²-Bench Telecom (AA)not in index | 45.6%#203 | 29.2%#256 | Artificial Analysis |
Knowledge
| Benchmark | thinking | default | Source |
|---|---|---|---|
| AA-Omniscience | -52.5#337 | -63.4#411 | Artificial Analysis |
Instruction following
| Benchmark | thinking | default | Source |
|---|---|---|---|
| IFBench | 59.4%#127 | 39.2%#259 | Artificial Analysis |
Multimodal
| Benchmark | thinking | default | Source |
|---|---|---|---|
| MMMU-Pro | 63.4%#156 | 64.3%#152 | Artificial Analysis |
Long context
| Benchmark | thinking | default | Source |
|---|---|---|---|
| AA-LCR | 55.3%#233 | – | Artificial Analysis |
Composite
| Benchmark | thinking | default | Source |
|---|---|---|---|
| AA Intelligence Index | 11.9#274 | 8.4#361 | Artificial Analysis |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Alibaba Cloud Int. | 51 tok/s | 1.02 s | $0.104 | $0.416 | 131k | – |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0009 | – | 28.2 s |
| Summarise a 30-page report | 12,000 / 600 | $0.0036 | – | 31.6 s |
| Code edit | 6,000 / 1,500 | $0.0053 | – | 41.6 s |
| Agentic coding session | 60,000 / 4,000 | $0.021 | – | 1.2 min |
| Structured extraction | 2,000 / 200 | $0.0009 | – | 27.1 s |
See also
Data as of 9 Sept 2026. Compare these configurations.