Alibaba
availableShows if the model has enough results for an index.Qwen3.6-27B
Qwen3.6-27B is a reasoning model from Alibaba. 48 benchmarks count toward its score, in 8 categories.
IndexOverall score. 50 is the middle.53.8 ±2.3
CoverageShare of the index weight with results.100%
SpeedOutput tokens per second.22/s
Input / 1MUS dollars per 1M input tokens.$0.3
Output / 1MUS dollars per 1M output tokens.$2
ContextMaximum tokens in one request.262K
EloLMArena rating and rank.N/A
50 is the middle of the board. The range shows the doubt in the index.
CapabilitiesScore per category. 50 is the middle.
50 is the middleResults
48 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. |
|---|---|---|---|---|---|---|
| CountBench | Multimodal | 97.8% | — | — | — | Qwen |
| V* | Multimodal | 94.7% | 54.9 | — | — | Z.AI |
| τ²-Bench Tool-Agent-User Evaluation | Agentic | 94.2% | 66.4 | — | — | Victor Barres et al. |
| AIME 2026 | Math | 94.1% | 53.5 | — | — | Qwen |
| Harvard-MIT Mathematics Tournament February 2025 | Math | 93.8% | 52.4 | — | — | Qwen |
| MMLU-Redux | Knowledge | 93.5% | 51.6 | — | — | Qwen |
| RefCOCO average | Multimodal | 92.5% | — | — | — | RefCOCO dataset authors |
| C-Eval | Knowledge | 91.4% | — | — | — | C-Eval authors |
| Harvard-MIT Mathematics Tournament November 2025 | Math | 90.7% | — | — | — | Qwen |
| Graduate-Level Google-Proof Q&A | Knowledge | 87.8% | 59.3 | — | — | David Rein et al. |
| Video-MME with subtitle | Multimodal | 87.7% | — | — | — | Qwen |
| MLVU mean average | Multimodal | 86.6% | — | — | — | Qwen |
| Massive Multitask Language Understanding Professional | Knowledge | 86.2% | 56.4 | — | — | Yubo Wang et al. |
| DynaMath | Multimodal | 85.6% | — | — | — | Qwen |
| GPQA diamond | Knowledge | 84.8% | 56.6 | none effort | — | Epoch AI |
| VideoMMMU | Multimodal | 84.4% | — | — | — | Qwen |
| Harvard-MIT Mathematics Tournament February 2026 | Math | 84.3% | 53.2 | — | — | Qwen |
| Artificial Analysis GPQA Diamond | Knowledge | 84.2% | 54.8 | — | — | Artificial Analysis |
| RealWorldQA | Multimodal | 84.1% | 54.5 | — | — | Qwen |
| LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code | Coding | 83.9% | 60.6 | — | — | Naman Jain et al. |
| Massive Multi-discipline Multimodal Understanding | Multimodal | 82.9% | 52.1 | — | — | MMMU authors |
| MStar | Multimodal | 81.4% | — | — | — | Qwen |
| CC-OCR | Multimodal | 81.2% | — | — | — | Qwen |
| MMAnswerBench | Math | 80.8% | — | — | — | Qwen |
| LiveBench Mathematics | Math | 79.9% | 54.6 | — | 25 Jun 2026 | LiveBench |
| CharXiv Reasoning | Multimodal | 78.4% | 55.3 | — | — | CharXiv authors |
| Artificial Analysis Long Context Reasoning | Reasoning | 77.3% | 61.7 | — | — | Artificial Analysis |
| Software Engineering Benchmark Verified | Coding | 77.2% | 59.5 | — | — | Carlos E. Jimenez et al. |
| Massive Multi-discipline Multimodal Understanding Pro | Multimodal | 75.8% | 50.7 | — | — | MMMU-Pro authors |
| Artificial Analysis MMMU-Pro | Multimodal | 74.6% | 58.3 | — | — | Artificial Analysis |
| Claw-Eval | Agentic | 72.4% | 72.2 | — | — | Bowen Ye et al. |
| LiveBench Coding | Coding | 71.8% | 56.9 | — | 25 Jun 2026 | LiveBench |
| LiveBench Data Analysis | Reasoning | 70.4% | 53.8 | — | 25 Jun 2026 | LiveBench |
| AndroidWorld | Agentic | 70.3% | — | — | — | Z.AI |
| LiveBench Reasoning | Reasoning | 70.3% | 53.6 | — | 25 Jun 2026 | LiveBench |
| SWE-bench | Coding | 70.0% | 53.7 | — | 1 Sept 2026 | Vals AI |
| Artificial Analysis IFBench | Instruction | 67.6% | 57.7 | — | — | Artificial Analysis |
| OTIS Mock AIME 2024-2025 | Math | 66.7% | 48.8 | none effort | — | Epoch AI |
| SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines | Knowledge | 66.0% | 52.1 | — | — | Xiaoxuan Du et al. |
| EuroEval Polish | Multilingual | 65.3% | 83.9 | — | — | EuroEval |
| LiveBench Language | Knowledge | 63.3% | 47.0 | — | 25 Jun 2026 | LiveBench |
| ERQA | Multimodal | 62.5% | 57.3 | — | — | Qwen |
| SimpleVQA | Multimodal | 56.1% | 45.9 | — | — | Z.AI |
| Gert Labs Composite Game Benchmark | Agentic | 54.8% | 61.2 | — | — | Gert Labs |
| Artificial Analysis Coding Index | Coding | 53.7% | 56.8 | — | — | Artificial Analysis |
| SWE-bench Pro | Coding | 53.5% | 55.7 | — | — | Xiang Deng et al. |
| QwenClawBench | Agentic | 53.4% | 52.8 | — | — | Qwen |
| LiveBench Instruction Following | Instruction | 53.2% | 44.3 | — | 25 Jun 2026 | LiveBench |
| Terminal-Bench 2.0 | Agentic | 44.9% | 54.8 | — | 4 Jun 2026 | Vals AI |
| Artificial Analysis SciCode | Coding | 42.8% | 52.3 | — | — | Artificial Analysis |
| LiveBench Agentic Coding | Agentic | 39.3% | 53.7 | — | 25 Jun 2026 | LiveBench |
| NL2Repo | Coding | 36.2% | 53.2 | — | — | MiniMax |
| FrontierMath-Tiers-1-3-v2-Private | Math | 34.0% | 50.6 | none effort | — | Epoch AI |
| Humanity's Last Exam | Knowledge | 24.0% | 49.1 | — | — | Center for AI Safety et al. |
| GDPval-AA normalized | Agentic | 23.7% | 53.8 | — | — | Artificial Analysis |
| Artificial Analysis Humanity's Last Exam | Knowledge | 23.1% | 49.6 | — | — | Artificial Analysis |
| Artificial Analysis Intelligence Index | Knowledge | 21.4% | 49.2 | — | — | Artificial Analysis |
| Artificial Analysis Agentic Index | Agentic | 20.1% | 53.5 | — | — | Artificial Analysis |
| Artificial Analysis Omniscience Accuracy | Knowledge | 19.6% | 38.1 | — | — | Artificial Analysis |
| Vibe Code Bench v1.1 | Coding | 11.9% | 47.1 | OpenHands | 21 Sept 2026 | Vals AI |
| Chess Puzzles | Reasoning | 9.0% | 35.2 | none effort | — | Epoch AI |
| Mystery Game Puzzles | Reasoning | 7.0% | 41.4 | none effort | — | Epoch AI |
| Critical Physics Tasks | Reasoning | 1.1% | 40.7 | — | — | Artificial Analysis |
48 benchmarks count, from 51 of 63 results. A grey row does not count. Too few models took that benchmark.