# HealthBench raw score (HealthBench (raw))

> Raw score on realistic multi-turn healthcare conversations graded against expert-written rubrics.

Canonical page: https://benchlm.ai/benchmarks/healthbench

- Category: [Knowledge](/knowledge)
- Last updated: September 27, 2026

## About HealthBench (raw)

- Year: 2026
- Tasks: 5,000 multi-turn patient conversations
- Format: Raw rubric score
- Difficulty: Realistic healthcare conversations
- Paper: [Claude Opus 5 System Card](https://www-cdn.anthropic.com/c5fbac3f0b1280a933ebd26d3cb8bb9f5bdeaf48/Claude%20Opus%205%20System%20Card.pdf)

Section 8.15.1 reports the unadjusted HealthBench score. This provider-run result is display-only and remains separate from the length-adjusted score.

HealthBench (raw) is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (5 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [Claude Opus 5.5](/models/claude-opus-5-5) | Anthropic | 68.1% |
| 2 | [Claude Opus 5](/models/claude-opus-5) | Anthropic | 67.1% |
| 3 | [GPT-6 Astra](/models/gpt-6-astra) | OpenAI | 56.9% |
| 4 | [GPT-6 Luna](/models/gpt-6-luna) | OpenAI | 50.0% |
| 5 | [GPT-6 Sol](/models/gpt-6-sol) | OpenAI | 47.1% |

## FAQ

### What does HealthBench (raw) measure?

Raw score on realistic multi-turn healthcare conversations graded against expert-written rubrics.

### Which model scores highest on HealthBench (raw)?

Claude Opus 5.5 by Anthropic currently leads with a score of 68.1% on HealthBench (raw).

### How many models are evaluated on HealthBench (raw)?

5 AI models have been evaluated on HealthBench (raw) on BenchLM.

### Does HealthBench (raw) affect BenchLM's overall score?

Not directly. HealthBench (raw) is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Compare Top Models on HealthBench (raw)

- [Claude Opus 5.5 vs Claude Opus 5](/compare/claude-opus-5-vs-claude-opus-5-5)
- [Claude Opus 5 vs GPT-6 Astra](/compare/claude-opus-5-vs-gpt-6-astra)
- [GPT-6 Astra vs GPT-6 Luna](/compare/gpt-6-astra-vs-gpt-6-luna)
- [GPT-6 Luna vs GPT-6 Sol](/compare/gpt-6-luna-vs-gpt-6-sol)