Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Reasoning benchmark
Anthropic STEM Professor Bench. Every model the catalog carries a published STEM Professor Bench value for, ranked by that value.
Anthropic's internal set of unpublished research problems written by professional scientists, each about a month of research work, mostly in theoretical physics; BenchLM stores the 64-problem subset. 3 models in the ModelScale catalog have a published value.
Each point is a model with a published STEM Professor Bench value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
3 models with a published STEM Professor Bench value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Mean score over up to nine attempts |
|---|---|---|---|
| 1 | Anthropic | 61.8 | |
| 2 | Anthropic | 53.7 | |
| 3 | Anthropic | 29.9 |
What the benchmark measures and how its value is published, from the source’s own record.
STEM Professor Bench (Anthropic STEM Professor Bench) is a reasoning benchmark published by Claude Fable 5.2 and Claude Mythos 5.2 system card. Anthropic's internal set of unpublished research problems written by professional scientists, each about a month of research work, mostly in theoretical physics; BenchLM stores the 64-problem subset. Its evaluation set is described by the source as 64 research problems. The source rates its difficulty as research-level science.
Each model's result is published by Claude Fable 5.2 and Claude Mythos 5.2 system card as mean score over up to nine attempts. ModelScale prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 3 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
Evidence key: Observed
Rows are ordered by the value Claude Fable 5.2 and Claude Mythos 5.2 system card published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.