Skip to content
View original post on X: Sara Hooker
Published
36/100

AutoScientist lifts open-model medical reasoning accuracy to 75.2%

AIAdaption says AutoScientist raised correctness on medical long-horizon reasoning from 58.1% to 75.2%, beating GPT-6 Astra and Gemini 4 Argon. The system works on any open-weights starting model and needs no retraining.
Sara HookerVerified on X
@sarahookr
View on X

AutoScientist increased correctness from 58.1% to 75.2% on medical long-horizon reasoning, outperforming GPT-6 Astra, Gemini 4 Argon. ✨

@adaption_ai supercharged performance on any open weights model for the task. 🔥 So you can use any starting model.

adaption@adaption_ai
New Adaption research: with AutoScientist, a lower-cost model outperformed GPT-6 Astra and Gemini 4 Argon on long medical records, at up to 15.4x lower cost. Correct answers jumped from 58% to 75%. No retraining.
View quoted post on X
Source details

Sara Hooker · x.comPublished · added here

Original title: AutoScientist increased correctness from 58.1% to 75.2% on medical long-horizon reasoning, outperforming GPT-6 Astra, Gemini 4 Argon. ✨

Topics

Open-source ecosystemModel releasesEvals & benchmarks