Best cell per row highlighted. Null means undisclosed — never counted as zero.
| Provider | Alibaba Cloud | Alibaba Cloud | Alibaba Cloud |
| Family / tier | Qwen3 · Large | Qwen3 · Large | Qwen3 · Medium |
| Status | GA | GA | GA |
| Released | 2025-04-28 | 2025-04-28 | 2025-04-28 |
| Knowledge cutoff | 2024-10 | 2024-10 | 2024-10 |
| Context window | 131K | 131K | 131K |
| Max output tokens | 33K | 33K | 33K |
| Input modalities | text | text | text |
| Reasoning mode | optional | optional | optional |
| Open weights | Yes | Yes | Yes |
| Input | $0.2 | $0.08 | $0.06 |
| Output | $0.6 | $0.28 | $0.2 |
| Cached input | — | — | — |
| Batch input | — | — | — |
| Free tier | Yes | Yes | Yes |
| Output speed (tok/s) | 68 | — | — |
| Time to first token (s) | 0.78 | — | — |
| Latency tier | medium | medium | fast |
| Trains on inputs | No | No | No |
| License | Apache-2.0 | Apache-2.0 | Apache-2.0 |
| Clouds | GCP | GCP | GCP |
| Compliance | — | — | — |
| Unified score | 8.5 | 8.4 | 7.8 |
| Decision Maker | 8 | 8.5 | 7.5 |
| Domain Strategist | 8.5 | 8 | 7.5 |
| Finance Lead | 9.5 | 9.5 | 9.5 |
| Domain Practitioner | 9 | 9 | 9 |
| Power User | 7.5 | 7.5 | 7 |
| Skeptic | 7 | 7.5 | 7.5 |
| Value score | 9.5 | 9.5 | 9.5 |
| MMLU-Pro | 82.8 | 65.54 | — |
| GPQA Diamond | 70 | — | — |
| AIME 2025 | 81.5 | — | — |
| LiveCodeBench | 70.7 | — | — |
| LMArena Elo | 1431 | — | — |
Scores link to their sources. Missing cells mean the vendor hasn't disclosed a result — honesty over padding.