Alibaba
availableShows if the model has enough results for an index.Qwen3.8-27B
Qwen3.8-27B is a reasoning model from Alibaba. 47 benchmarks count toward its score, in 8 categories.
IndexOverall score. 50 is the middle.63.1 ±2.4
CoverageShare of the index weight with results.100%
SpeedOutput tokens per second.44/s
Input / 1MUS dollars per 1M input tokens.$0.42
Output / 1MUS dollars per 1M output tokens.$3
ContextMaximum tokens in one request.1M
EloLMArena rating and rank.1439 (#68)
50 is the middle of the board. The range shows the doubt in the index.
10,697 votes. Elo shows what people prefer. It does not change the score.
CapabilitiesScore per category. 50 is the middle.
50 is the middleResults
47 counted| BenchmarkThe test name. | CategoryThe capability that the test measures. | ResultThe score from the publisher. | IndexThis result on the index scale. | RunThe settings of the run. | DateDate of the result. | Published byThe source of the result. |
|---|---|---|---|---|---|---|
| MathVision with Python | Multimodal | 94.6% | — | — | — | Moonshot AI / MathVision authors |
| OmniDocBench 1.5 | Multimodal | 91.1% | — | — | — | OpenAI |
| Artificial Analysis GPQA Diamond | Knowledge | 90.5% | 61.3 | — | — | Artificial Analysis |
| LiveCodeBench v6 | Coding | 90.3% | 59.2 | — | — | LiveCodeBench maintainers |
| CharXiv Reasoning | Multimodal | 90.2% | 68.8 | — | — | CharXiv authors |
| MathVision | Multimodal | 90.0% | — | — | — | Qwen |
| Graduate-Level Google-Proof Q&A | Knowledge | 89.2% | 60.6 | — | — | David Rein et al. |
| GPQA Diamond | Knowledge | 89.2% | 60.6 | — | — | David Rein et al. |
| GPQA Diamond | Knowledge | 88.9% | 60.3 | — | 1 Sept 2026 | Vals AI |
| LiveBench Mathematics | Math | 86.2% | 63.1 | — | 25 Jun 2026 | LiveBench |
| SWE-bench | Coding | 86.0% | 66.5 | — | 1 Sept 2026 | Vals AI |
| RealWorldQA | Multimodal | 85.9% | 58.0 | — | — | Qwen |
| BabyVision with Python | Multimodal | 85.6% | — | — | — | Moonshot AI |
| MMLU Pro | Knowledge | 84.3% | 53.4 | — | 1 Sept 2026 | Vals AI |
| OSWorld-Verified | Agentic | 84.3% | 72.2 | — | — | Tianbao Xie et al. |
| LiveCodeBench | Coding | 84.0% | 60.7 | — | 1 Sept 2026 | Vals AI |
| MMMU Pro | Multimodal | 83.9% | 63.9 | — | 1 Sept 2026 | Vals AI |
| CharXiv Reasoning without tools | Multimodal | 83.7% | — | — | — | CharXiv authors |
| VulcanBench v3 | Coding | 82.6% | 65.6 | — | — | VulcanBench contributors |
| Artificial Analysis Long Context Reasoning | Reasoning | 82.0% | 65.0 | — | — | Artificial Analysis |
| AndroidWorld | Agentic | 81.9% | — | — | — | Z.AI |
| LiveBench Reasoning | Reasoning | 80.0% | 67.1 | — | 25 Jun 2026 | LiveBench |
| Instruction Following Benchmark | Instruction | 79.5% | 58.5 | — | — | Benchmark authors |
| LiveBench Data Analysis | Reasoning | 76.6% | 62.3 | — | 25 Jun 2026 | LiveBench |
| Artificial Analysis MMMU-Pro | Multimodal | 76.3% | 60.3 | — | — | Artificial Analysis |
| LiveBench Coding | Coding | 75.7% | 63.3 | — | 25 Jun 2026 | LiveBench |
| LiveBench Language | Knowledge | 74.3% | 60.3 | — | 25 Jun 2026 | LiveBench |
| Terminal-Bench 2.1 (provider run) | Agentic | 73.0% | 67.1 | — | — | DeepSeek-AI |
| Terminal-Bench 2.1 (provider run) | Agentic | 73.0% | 67.1 | — | — | DeepSeek-AI |
| LiveBench Instruction Following | Instruction | 72.7% | 74.7 | — | 25 Jun 2026 | LiveBench |
| CoWorkBench | Agentic | 70.7% | — | — | — | Qwen Team |
| EuroEval Portuguese | Multilingual | 68.5% | 87.8 | — | — | EuroEval |
| Artificial Analysis Coding Index | Coding | 68.1% | 66.9 | — | — | Artificial Analysis |
| EuroEval French | Multilingual | 66.8% | 85.7 | — | — | EuroEval |
| EuroEval Italian | Multilingual | 66.2% | 84.9 | — | — | EuroEval |
| BabyVision | Multimodal | 65.7% | — | — | — | Meta AI |
| ERQA | Multimodal | 65.5% | 60.1 | — | — | Qwen |
| EuroEval Polish | Multilingual | 65.4% | 83.9 | — | — | EuroEval |
| EuroEval Swedish | Multilingual | 64.9% | 83.3 | — | — | EuroEval |
| Vibe Code Bench v1.1 | Coding | 64.8% | 69.1 | OpenHands | 21 Sept 2026 | Vals AI |
| WebArena-Verified Browser Agent Benchmark | Agentic | 64.8% | — | — | — | Amine El Hattami et al. |
| Vision2Web | Multimodal | 62.9% | — | — | — | Z.AI |
| EuroEval Dutch | Multilingual | 62.1% | 79.8 | — | — | EuroEval |
| SWE-bench Pro | Coding | 61.7% | 63.7 | — | — | Xiang Deng et al. |
| LiveBench Agentic Coding | Agentic | 61.4% | 74.5 | — | 25 Jun 2026 | LiveBench |
| EuroEval Spanish | Multilingual | 59.0% | 76.0 | — | — | EuroEval |
| Terminal-Bench 2.1 | Agentic | 58.4% | 58.5 | — | 21 Sept 2026 | Vals AI |
| EuroEval German | Multilingual | 57.2% | 73.6 | — | — | EuroEval |
| Artificial Analysis Tau3-Banking | Agentic | 48.0% | 77.8 | — | — | Artificial Analysis |
| Artificial Analysis SciCode | Coding | 46.6% | 57.5 | — | — | Artificial Analysis |
| Artificial Analysis Agentic Index | Agentic | 46.5% | 75.1 | — | — | Artificial Analysis |
| GDPval-AA normalized | Agentic | 45.4% | 70.4 | — | — | Artificial Analysis |
| Artificial Analysis EnterpriseOps-Gym | Agentic | 44.2% | 67.9 | — | — | Artificial Analysis |
| Agents' Last Exam | Agentic | 42.9% | 81.6 | — | — | DeepSeek-AI |
| NL2Repo | Coding | 42.3% | 58.1 | — | — | MiniMax |
| DeepSWE | Agentic | 42.2% | 54.5 | — | — | Datacurve AI |
| IOI | Coding | 39.1% | 62.7 | — | 21 Sept 2026 | Vals AI |
| SkillsBench | Coding | 38.1% | 55.3 | OpenHands | 11 Sept 2026 | Vals AI |
| Artificial Analysis Humanity's Last Exam | Knowledge | 33.9% | 61.3 | — | — | Artificial Analysis |
| Artificial Analysis Intelligence Index | Knowledge | 33.7% | 64.5 | — | — | Artificial Analysis |
| JobBench | Agentic | 33.4% | 57.1 | — | — | Yuetai Li et al. |
| Humanity's Last Exam | Knowledge | 30.8% | 54.9 | — | — | Center for AI Safety et al. |
| Humanity's Last Exam without tools | Knowledge | 30.8% | 54.9 | — | — | OpenAI |
| Medical Long Context Reasoning (MLCR-AA) | Reasoning | 21.7% | 56.6 | — | — | Wisedocs and Artificial Analysis |
| ProofBench v1.1 | Math | 16.0% | 55.9 | — | 21 Sept 2026 | Vals AI |
| Artificial Analysis Omniscience Accuracy | Knowledge | 15.6% | 33.2 | — | — | Artificial Analysis |
| Code Migration | Coding | 14.2% | 54.0 | — | 21 Sept 2026 | Vals AI |
| Critical Physics Tasks | Reasoning | 5.4% | 49.7 | — | — | Artificial Analysis |
| Agent Arena task outcome | Agentic | 4.3 | 72.2 | — | 15 Sept 2026 | LMArena |
| Terminal-Bench 4.0 | Agentic | 4.0% | 62.3 | — | 21 Sept 2026 | Vals AI |
| Agent Arena steerability | Agentic | 0.1 | 67.5 | — | 15 Sept 2026 | LMArena |
| ProgramBench | Coding | 0.0% | — | — | 21 Sept 2026 | Vals AI |
| Agent Arena command recovery | Agentic | -5.2 | 61.5 | — | 15 Sept 2026 | LMArena |
47 benchmarks count, from 62 of 73 results. A grey row does not count. Too few models took that benchmark.