MAST (Medical AI Superintelligence Test)
A composite of clinical benchmarks spanning diagnostic and management reasoning, safety, multimodal imaging, and agentic capability.
Preview: MAST is currently in preview. Exact scores on this benchmark may change as we undergo final validation and tuning.
More about this boardLess
Composite score across curated clinical benchmarks spanning diagnostic reasoning, management reasoning, safety, multimodal images, multimodal radiology, and agentic capability. Components: First Do NOHARM v2, SCT-Bench, MedAgentBench v2, PhysicianBench, ReXrank Mini, CPC-Bench.
The board is marked as a preview and was last updated August 15, 2026; component-level breakdowns are published only for First, Do NOHARM v2. Scores may move before the full release.
Published by ARISE AI Research Network (multi-institutional), released Aug 2026. Composite of 6 component benchmarks; 11 models. Percentage composite, higher better.
- Board marked as preview
- Yes
- How the composite is formed
- harmonic mean of Diagnostic Reasoning, Management Reasoning, Safety, Multimodal Radiology, Multimodal Images
- Board last updated
- 15 Aug 2026
- Rows the board shows by default
- latest large and small model from each provider
- Board generated
- 2026-08-15 18:28 UTC
- Models in the board's default view
- 11
- Publisher
- ARISE AI Research Network (multi-institutional)
- Rows
- 8
- Models
- 8 of 11 on the board
- Released
- Aug 2026
- Measured to
- Aug 2026
- Unit
- 6-benchmark composite
- Documents cited
- 1
Every result on MAST (Medical AI Superintelligence Test), on the board's own scale
8 results, one column each, highest first. Percentage composite, higher better.
Column colour is the model's lab. The table below says who ran each result.
Every published result, ranked on this board alone
8 rows. Higher is better. Bars run from 0 to 100; whiskers show a published interval. Select a row for its quote, page and document.
1GPT-5.6 SolOpenAI · MAST in preview60.2Board publisherAug 2026[1]
1 GPT-5.6 Sol OpenAI 60.2%
2Kimi K3Moonshot AI60.1Board publisherAug 2026[1]
2 Kimi K3 Moonshot AI 60.1%
3Gemini 3.6 FlashGoogle59.3Board publisherAug 2026[1]
3 Gemini 3.6 Flash Google 59.3%
4Gemini 3.1 ProGoogle58.9Board publisherAug 2026[1]
4 Gemini 3.1 Pro Google 58.9%
5Qwen3.5 397B A17BAlibaba57.9Board publisherAug 2026[1]
5 Qwen3.5 397B A17B Alibaba 57.9%
6Claude Opus 5Anthropic57.1Board publisherAug 2026[1]
6 Claude Opus 5 Anthropic 57.1%
7Claude Sonnet 5Anthropic56.6Board publisherAug 2026[1]
7 Claude Sonnet 5 Anthropic 56.6%
8Grok 4.3SpaceX AI53.7Board publisherAug 2026[1]
8 Grok 4.3 xAI 53.7%
1 document supports this board
- 1.MAST: Medical AI Superintelligence Test leaderboard (General board)official leaderboard, ARISE AI Research Network, 15 Aug 2026. Every result it supports