MAI-Thinking-1 on HealthBench Professional
MAI-Thinking-1 scores 0.350 on HealthBench Professional, rank 42 of 43 models on the board. Microsoft AI's first in-house frontier reasoning model, a sparse mixture-of-experts design in public preview on Microsoft Foundry. HealthBench Professional scores models on 525 tasks drawn from real clinician conversations, graded criterion by criterion against physician-written rubrics on a 0 to 1 scale.
Where it sits on the board
The models ranked around MAI-Thinking-1, by length-adjusted score. Differences are against MAI-Thinking-1.
| Rank | Model | Length-adjusted | Before the adjustment | Difference |
|---|---|---|---|---|
| 37 | 0.459 | 0.500 | +0.109 | |
| 38 | 0.442 | – | +0.092 | |
| 39 | 0.441 | 0.468 | +0.091 | |
| 40 | 0.396 | 0.480 | +0.046 | |
| 41 | 0.384 | 0.407 | +0.034 | |
| 42 | 0.350 | – | – | |
| 43 | 0.322 | – | -0.028 |
Result and API facts
| Rank | 42 of 43 |
|---|---|
| Length-adjusted score | 0.350 |
| Score before the length adjustment | not printed in the source |
| Lab | Microsoft |
| Context window | 256K tokens |
| API price per 1M tokens | $2.00 in / $8.00 out |
| License | proprietary |
| Source | MAI-Thinking-1: Building a Hill-Climbing Machine (model card) |
| Parameters | 1.0T |
| Released | 2026-08-12 |
Source of this score
Read from MAI-Thinking-1: Building a Hill-Climbing Machine (model card, Microsoft AI, 2026-08-12). Vendor-reported. Confidence: verified. Configuration: length-adjusted (HealthBench Professional length penalty), standard GPT-5.4 grader and OpenAI rubrics, Microsoft AI run; printed at integer precision..
Model AIR-Bench CyberSec Instruct CyberSec Auto Long Fact Truthful QA HealthBench Prof. MedXpert QA MAI-Thinking-1 88 63 63 98 88 35 43 Sonnet 4.6 88 62 56 98 88 38 49
Position in the field
The gap to the leader, GPT-6 Astra (Anthropic run) at 0.703, is 0.353. Directly above sits GPT-5.5 Instant at 0.384. Directly below sits Claude Haiku 4.5 at 0.322. The scores on this page are compiled from published documents rather than from one controlled run, so a small gap between 2 models can reflect a difference in grader version, reasoning effort, or deployment setting as well as a difference in capability.
What does MAI-Thinking-1 score on HealthBench Professional?
MAI-Thinking-1 scores 0.350 on HealthBench Professional, which places it at rank 42 of 43 models on the board as of November 2, 2026. The number was read from MAI-Thinking-1: Building a Hill-Climbing Machine, listed on the sources page.
How much does MAI-Thinking-1 cost to run?
MAI-Thinking-1 is priced at $2.00 per million input tokens and $8.00 per million output tokens through Microsoft's API.
Head to head
Score-difference matrixPairings with a dedicated comparison page are linked; every other difference is in the score-difference matrix.
- MAI-Thinking-1 vs GPT-6 Astra (Anthropic run)
- MAI-Thinking-1 vs Claude Sonnet 5.5
- MAI-Thinking-1 vs Claude Fable 5
- MAI-Thinking-1 vs Claude Opus 5.5
- MAI-Thinking-1 vs GPT-6 Luna (Anthropic run)
- MAI-Thinking-1 vs Claude Haiku 5.5
- MAI-Thinking-1 vs GPT-6 Astra
- MAI-Thinking-1 vs GPT-6.1 Sol
- MAI-Thinking-1 vs Claude Fable 5 (September card)
- MAI-Thinking-1 vs GPT-6.1 Terra
- MAI-Thinking-1 vs Muse Wonder Full
- MAI-Thinking-1 vs Claude Fable 5.1
- MAI-Thinking-1 vs GPT-6 Luna
- MAI-Thinking-1 vs GPT-6 Sol
- MAI-Thinking-1 vs GPT-5.6 Sol
- MAI-Thinking-1 vs Muse Wonder
- MAI-Thinking-1 vs Claude Opus 5
- MAI-Thinking-1 vs Muse Spark 1.1
- MAI-Thinking-1 vs Claude Sonnet 5
- MAI-Thinking-1 vs GPT-5.6 Terra
- MAI-Thinking-1 vs Claude Opus 4.8 (Opus 4.8 grader)
- MAI-Thinking-1 vs Grok 4.7
- MAI-Thinking-1 vs Claude Opus 4.8
- MAI-Thinking-1 vs GPT-5.6 Luna
- MAI-Thinking-1 vs Gemini 4 Argon
- MAI-Thinking-1 vs Muse Spark
- MAI-Thinking-1 vs GPT-5.6 Sol (August)
- MAI-Thinking-1 vs Kimi K3
- MAI-Thinking-1 vs GLM 5.3
- MAI-Thinking-1 vs Claude Opus 4.7
- MAI-Thinking-1 vs GPT-5.5
- MAI-Thinking-1 vs Gemini 3.8 Flash
- MAI-Thinking-1 vs DeepSeek V4.1 Flash
- MAI-Thinking-1 vs Grok 4.6
- MAI-Thinking-1 vs GPT-5.4
- MAI-Thinking-1 vs GPT-5
- MAI-Thinking-1 vs GPT-5.2
- MAI-Thinking-1 vs Claude Sonnet 4.6
- MAI-Thinking-1 vs GPT-5.6 Luna (August)
- MAI-Thinking-1 vs GPT-5.1
- MAI-Thinking-1 vs GPT-5.5 Instant
- MAI-Thinking-1 vs Claude Haiku 4.5
Where the scores come from and how they are read is on the methodology page. The full ranking is on the leaderboard.