Health AI Benchmarks

Claude Sonnet 5

4 results on 4 boards

Made by
Anthropic
Released
30 June 2026
Licence
proprietary
Context
1.0M
Input, per 1M tokens
$2
Output, per 1M tokens
$10

Health Evals Index

56.8rank 25 of 533 of 4 boards

This site's own calculation from 3 of 4 benchmarks. How it works

How this score is made
  • HealthBench Professional0.57857.8 out of 100runs −2.5 hardadjusted 60.3weight 0.446, supportingsystem cardReported by the maker.length-adjusted, Anthropic protocol: adaptive thinking at max effort, Claude Opus 4.8 grader, averaged over 5 trials, no tools or custom system prompt (raw 62.4%).
  • HealthBench58.758.7 out of 100runs −0.5 leveladjusted 59.2weight 0.798, coresystem cardReported by the maker.length-adjusted; Anthropic protocol: adaptive thinking at max effort, Claude Opus 4.8 grader, averaged over 5 trials, no tools or custom system prompt; figure-only in the Sonnet 5 card; raw 59.2% per Opus 5 card
  • Health Optimization Bench: incretin50.150.1 out of 100runs +4.0 easyadjusted 46.1weight 0.226, supportingofficial leaderboardNear its ceiling (leader 84.9), so it counts for less.
  • SAFE-TeenNo result here, and none is filled in.
Adjusted average of 3 benchmarks, by weight
57.5
Field average, counts for 33.3% with 3 of 4 benchmarks
55.3
Index
= 56.8

Runs: how far that benchmark's scores sit above or below the same models' scores on the others. Adjusted: the score minus that amount. The fewer benchmarks behind a score, the more the field average counts.

Benchmark by benchmark

Claude Sonnet 5 next to the top 10 on each benchmark it has a score on, on that benchmark's own scale. Its own column is in colour.

Every result, with its source

Each against its own board only. The dot strip shows the whole board; this model is the larger red dot.

Compare Claude Sonnet 5 with other models