Skip to main content

MAI-Thinking-1

MicrosoftCurrentReleased Jun 2, 2026
Overall Score
65Prov. #43 of 122Verified #23 of 31
Arena Elo
N/A
Categories Ranked
3of 8
Price (1M tokens)
N/A
Speed
N/A
Context
256K
ProprietaryReasoning
Confidence
1

According to BenchLM.ai, MAI-Thinking-1 ranks #43 out of 122 models on the provisional leaderboard with an overall score of 65/100. It also ranks #23 out of 31 on the verified leaderboard. While not a frontier model, it offers specific advantages depending on the use case.

MAI-Thinking-1 is a proprietary model with a 256K token context window. It uses explicit chain-of-thought reasoning, which typically improves performance on math and complex reasoning tasks at the cost of higher latency and token usage.

This profile currently has 14 of 236 tracked benchmarks. BenchLM only exposes non-generated benchmark rows publicly, so missing categories stay blank until a sourced evaluation is available.

Its strongest category is Instruction Following (#1), while its weakest is Knowledge (#47). This performance profile makes it a well-rounded choice across a range of tasks.

Ranking Distribution

Category rank across 5 benchmark categories — sorted by best rank

Category Performance

Scores across all benchmark categories (0-100 scale)

Category Breakdown

Agentic

31.9/ 100
Weight: 22%1 benchmark
Terminal-Bench 2.0BrowseCompOSWorld-VerifiedGAIATAU-benchWebArena

Coding

#15
84.5/ 100
Weight: 20%4 benchmarks
SWE-bench VerifiedLiveCodeBenchSWE-bench ProSWE-RebenchSciCode

Reasoning

0.0/ 100
Weight: 17%1 benchmark
MuSRLongBench v2MRCRv2ARC-AGI-2

Knowledge

#47
64.6/ 100
Weight: 12%4 benchmarks
GPQASuperGPQAMMLU-ProHLEFrontierScienceSimpleQA

Math

98.4/ 100
Weight: 5%3 benchmarks
AIME 2025BRUMO 2025MATH-500FrontierMath

Multilingual

0.0/ 100
Weight: 7%0 benchmarks
MGSMMMLU-ProX

Multimodal

0.0/ 100
Weight: 12%0 benchmarks
MMMU-ProOfficeQA ProCharXivCharXiv w/o tools

Inst. Following

#1
100.0/ 100
Weight: 5%1 benchmark
IFEvalIFBench

Benchmark Details

Only benchmark rows with an attached exact-source record are shown here. Source-unverified manual rows and generated rows are hidden from model pages.

Frequently Asked Questions

How does MAI-Thinking-1 perform overall in AI benchmarks?

MAI-Thinking-1 currently ranks #43 out of 122 models on BenchLM's provisional leaderboard with an overall score of 65. It also ranks #23 out of 31 on the verified leaderboard. It is created by Microsoft and features a 256K context window.

Is MAI-Thinking-1 good for knowledge and understanding?

MAI-Thinking-1 ranks #47 out of 122 models in knowledge and understanding benchmarks with an average score of 64.6. There are stronger options in this category.

Is MAI-Thinking-1 good for coding and programming?

MAI-Thinking-1 ranks #15 out of 122 models in coding and programming benchmarks with an average score of 84.5. There are stronger options in this category.

Is MAI-Thinking-1 good for mathematics?

MAI-Thinking-1 has visible benchmark coverage in mathematics, but BenchLM does not currently assign it a global category rank there.

Is MAI-Thinking-1 good for reasoning and logic?

MAI-Thinking-1 has visible benchmark coverage in reasoning and logic, but BenchLM does not currently assign it a global category rank there.

Is MAI-Thinking-1 good for agentic tool use and computer tasks?

MAI-Thinking-1 has visible benchmark coverage in agentic tool use and computer tasks, but BenchLM does not currently assign it a global category rank there.

Is MAI-Thinking-1 good for instruction following?

MAI-Thinking-1 ranks #1 out of 122 models in instruction following benchmarks with an average score of 100. It is among the top performers in this category.

Does MAI-Thinking-1 have full benchmark coverage on BenchLM?

Not yet. MAI-Thinking-1 currently has 14 published benchmark scores out of the 236 benchmarks BenchLM tracks. BenchLM only exposes non-generated public benchmark rows, so missing categories stay blank until a sourced evaluation is available.

What is the context window size of MAI-Thinking-1?

MAI-Thinking-1 has a context window of 256K, which determines how much text it can process in a single interaction.

Last updated: June 8, 2026 · Runtime metrics stay blank until BenchLM has a sourced snapshot.

Don't miss the next GPT moment

Which models moved up, what’s new, and what it costs. One email a week, 3-min read.

Free. One email per week.