Multimodal & Grounded benchmark
Biomedical image analysis leaderboard
Anthropic biomedical-image-analysis evaluation. Every model the catalog carries a published Biomedical image analysis value for, ranked by that value.
CategoryMultimodal & Grounded
MeasureNormalized task metric
Tasks91 cases across 31 biomedical imaging tasks
DifficultyBiomedical image analysis
Quantitative biomedical image analysis across segmentation, detection, measurement, classification, registration, reconstruction, and tracking tasks.
Biomedical image analysis ranking
2 models with a published Biomedical image analysis value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
Models ranked by their published Anthropic biomedical-image-analysis evaluation value| Rank | Model | Provider | Normalized task metric |
|---|
| 1 |  Claude Sonnet 5.5 | Anthropic | 72.2 |
| 2 | Claude Opus 5.5 | Anthropic | 71.4 |
Evidence key: Observed
Rows are ordered by the value Claude Opus 5.5 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.
About Biomedical image analysis
What the benchmark measures and how its value is published, from the source’s own record.
What is Biomedical image analysis?
Biomedical image analysis (Anthropic biomedical-image-analysis evaluation) is a multimodal & grounded benchmark published by Claude Opus 5.5 System Card. Quantitative biomedical image analysis across segmentation, detection, measurement, classification, registration, reconstruction, and tracking tasks. Its evaluation set is described by the source as 91 cases across 31 biomedical imaging tasks. The source rates its difficulty as biomedical image analysis.
How is Biomedical image analysis scored?
Each model's result is published by Claude Opus 5.5 System Card as normalized task metric. ModelScale prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 2 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.
ModelScale