Medcasereasoning
MedCaseReasoning medical diagnosis evaluation
- Domain
- rl-env
- License
- unknown
- Published
- Sep 2025
Cite
Notes
Only stored in your browser.
Top score 26.7% by GPT-5 Nano - 3 models reporting (1 frontier)
Score history
2Top models
3
Related tools
1Implementations, trainers, datasets and scaffolds linked to this eval.
FAQ
- What is Medcasereasoning?
- MedCaseReasoning medical diagnosis evaluation
- What is the current top score on Medcasereasoning?
- The top reported score is 26.7% by GPT-5 Nano, across 3 models reporting (1 from frontier labs).
- How can a model improve its Medcasereasoning score?
- Tools linked to Medcasereasoning on Sophon include Medcasereasoning RL Env (Medarc) - RL environments, datasets, and scaffolds that target this eval.
- What license is Medcasereasoning under?
- Medcasereasoning is available under unknown.