Health AI Benchmarks
Menu

Claude Haiku 5.5

2 results on 2 boards

Made by
Anthropic
Released
7 October 2026
Licence
proprietary
Context
1M
Input, per 1M tokens
$0.1
Output, per 1M tokens
$0.5

Health Evals Index

44.6rank 10 of 612 of 5 boards

This site's own calculation from 2 of 5 benchmarks, its weighted average times a multiplier: 63.1 × 0.707 = 44.6. How it works

  • HealthBench Professional0.64864.8weight 0.743system cardReported by the maker.Anthropic; max effort; Opus 4.8 grader; safety classifiers enabled; length-adjusted; raw 71.0%; five runs (October 1 and 2); thirteen repeat max runs on October 4 averaged 64.2%
  • HealthBench61.661.6weight 0.823system cardReported by the maker.Anthropic; max effort; Opus 4.8 grader; safety classifiers enabled; length-adjusted; raw 64.2%; mean of five runs made on October 2
  • Health Optimization Benchno result on this benchmark
  • HealthBench Hardno result on this benchmark
  • WHBenchno result on this benchmark

63.1 × 0.707 = 44.6: the weighted average of these 2 scores out of 100, times 0.707 because this score rests on 2 of 5 benchmarks. From 4 benchmarks up a score keeps its full average. Each benchmark counts by its weight, which is lower the nearer its leader is to 100.

Every result

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

  • HealthBench Professional

    Anthropic; max effort; Opus 4.8 grader; safety classifiers enabled; length-adjusted; raw 71.0%; five runs (October 1 and 2); thirteen repeat max runs on October 4 averaged 64.2%

    0.648

    6 of 35, leader 0.703

    system card Oct 2026

  • HealthBench

    Anthropic; max effort; Opus 4.8 grader; safety classifiers enabled; length-adjusted; raw 64.2%; mean of five runs made on October 2

    61.6

    6 of 31, leader 67.1

    system card Oct 2026

Compare Claude Haiku 5.5 with other models