Alibaba
availableShows if the model has enough results for an index.Qwen3 Max
Qwen3 Max is a reasoning model from Alibaba. 23 benchmarks count toward its score, in 6 categories.
IndexOverall score. 50 is the middle.44.5 ±4.9
CoverageShare of the index weight with results.85%
SpeedOutput tokens per second.34/s
Input / 1MUS dollars per 1M input tokens.$0.78
Output / 1MUS dollars per 1M output tokens.$3.9
ContextMaximum tokens in one request.262K
EloLMArena rating and rank.1413 (#124)
50 is the middle of the board. The range shows the doubt in the index.
8,939 votes. Elo shows what people prefer. It does not change the score.
CapabilitiesScore per category. 50 is the middle.
50 is the middleResults
23 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. |
|---|---|---|---|---|---|---|
| MATH level 5 | Math | 97.1% | 46.4 | — | — | Epoch AI |
| MGSM | Multilingual | 92.1% | — | — | 9 Jan 2026 | Vals AI |
| MGSM | Multilingual | 91.8% | — | — | 9 Jan 2026 | Vals AI |
| MMLU Pro | Knowledge | 84.7% | 53.9 | — | 1 Sept 2026 | Vals AI |
| τ²-bench Telecom | Agentic | 84.2% | 59.2 | Alibaba | 2 Mar 2026 | Sierra Research |
| MMLU Pro | Knowledge | 83.5% | 52.1 | — | 1 Sept 2026 | Vals AI |
| GPQA Diamond | Knowledge | 82.2% | 54.1 | — | 1 Sept 2026 | Vals AI |
| AIME | Math | 81.0% | 51.5 | — | 16 Apr 2026 | Vals AI |
| LiveCodeBench | Coding | 78.2% | 55.4 | — | 1 Sept 2026 | Vals AI |
| GPQA Diamond | Knowledge | 77.8% | 50.0 | — | 1 Sept 2026 | Vals AI |
| Artificial Analysis GPQA Diamond | Knowledge | 76.4% | 46.9 | — | — | Artificial Analysis |
| τ²-Bench Tool-Agent-User Evaluation | Agentic | 74.3% | 52.1 | — | — | Victor Barres et al. |
| OTIS Mock AIME 2024-2025 | Math | 73.3% | 52.6 | — | — | Epoch AI |
| GPQA diamond | Knowledge | 72.6% | 45.3 | — | — | Epoch AI |
| τ²-bench Retail | Agentic | 72.2% | 50.6 | Alibaba | 30 Apr 2026 | Sierra Research |
| LiveCodeBench | Coding | 66.9% | 45.2 | — | 1 Sept 2026 | Vals AI |
| AIME | Math | 60.7% | 42.0 | — | 16 Apr 2026 | Vals AI |
| τ²-bench Airline | Agentic | 59.5% | 41.5 | Alibaba | 2 Mar 2026 | Sierra Research |
| Artificial Analysis Long Context Reasoning | Reasoning | 50.0% | 42.8 | — | — | Artificial Analysis |
| SimpleQA Verified | Knowledge | 48.7% | 66.7 | — | — | Epoch AI |
| Artificial Analysis IFBench | Instruction | 44.1% | 33.9 | — | — | Artificial Analysis |
| Gert Labs Composite Game Benchmark | Agentic | 43.7% | 51.4 | — | — | Gert Labs |
| Terminal-Bench 1.0 | Agentic | 36.3% | 42.3 | — | 12 Jan 2026 | Vals AI |
| Terminal-Bench 1.0 | Agentic | 36.3% | 42.3 | — | 12 Jan 2026 | Vals AI |
| Artificial Analysis Omniscience Accuracy | Knowledge | 24.4% | 44.1 | — | — | Artificial Analysis |
| Terminal-Bench 2.0 | Agentic | 22.5% | 38.7 | — | 4 Jun 2026 | Vals AI |
| FrontierMath-Tiers-1-3-v2-Private | Math | 18.9% | 42.1 | — | — | Epoch AI |
| Artificial Analysis Intelligence Index | Knowledge | 15.6% | 41.9 | — | — | Artificial Analysis |
| IOI v1 | Coding | 14.7% | 48.3 | — | 9 Aug 2026 | Vals AI |
| Artificial Analysis Humanity's Last Exam | Knowledge | 11.9% | 37.5 | — | — | Artificial Analysis |
| IOI v1 | Coding | 7.8% | 44.3 | — | 9 Aug 2026 | Vals AI |
| Mystery Game Puzzles | Reasoning | 5.0% | 39.2 | — | — | Epoch AI |
| Chess Puzzles | Reasoning | 4.0% | 28.7 | — | — | Epoch AI |
| Vibe Code Bench v1.1 | Coding | 3.5% | 43.6 | OpenHands | 21 Sept 2026 | Vals AI |
| Critical Physics Tasks | Reasoning | 0.0% | 38.4 | — | — | Artificial Analysis |
23 benchmarks count, from 33 of 35 results. A grey row does not count. Too few models took that benchmark.