Microsoft
availableShows if the model has enough results for an index.MAI-Thinking-1
MAI-Thinking-1 is a reasoning model from Microsoft in the MAI-Thinking family. 9 benchmarks count toward its score, in 4 categories.
IndexOverall score out of 100.55.1 ±10.4
CoverageShare of the index weight with results.50%
SpeedOutput tokens per second.—
Input / 1MUS dollars per 1M input tokens.N/A
Output / 1MUS dollars per 1M output tokens.N/A
ContextMaximum tokens in one request.256K
EloLMArena rating and rank.N/A
The index is a score out of 100. The ± range shows how much it can change.
CapabilitiesScore per category, out of 100.
Out of 100Results
9 counted| BenchmarkThe test name. | CategoryThe capability that the test measures. | ResultThe score from the publisher. | IndexThis result as a score out of 100. | RunThe settings of the run. | DateDate of the result. | Published byThe source of the result. |
|---|---|---|---|---|---|---|
| American Invitational Mathematics Examination 2025 | Math | 97.0% | 52.9 | — | — | Mathematical Association of America |
| AIME 2026 | Math | 94.5% | 53.8 | — | — | Qwen |
| Graphwalks BFS 0K-128K | Reasoning | 90.0% | — | — | — | OpenAI |
| LiveCodeBench v6 | Coding | 87.7% | 56.8 | — | — | LiveCodeBench maintainers |
| Massive Multitask Language Understanding Professional | Knowledge | 85.0% | 54.5 | — | — | Yubo Wang et al. |
| Instruction Following Benchmark | Instruction | 85.0% | 64.9 | — | — | Benchmark authors |
| Harvard-MIT Mathematics Tournament February 2026 | Math | 84.9% | 53.7 | — | — | Qwen |
| Graduate-Level Google-Proof Q&A | Knowledge | 84.2% | 56.0 | — | — | David Rein et al. |
| GPQA Diamond | Knowledge | 84.2% | 56.0 | — | — | David Rein et al. |
| Software Engineering Benchmark Verified | Coding | 73.5% | 56.5 | — | — | Carlos E. Jimenez et al. |
| SWE-bench Pro | Coding | 52.8% | 55.1 | — | — | Xiang Deng et al. |
| Measuring Short-Form Factuality in Large Language Models | Knowledge | 31.0% | — | — | — | Jason Wei et al. |
9 benchmarks count, from 10 of 12 results. A grey row does not count. Too few models took that benchmark.