# HealthBench (consensus) — AI model rankings

physician rubric score on the consensus subset of the original HealthBench evaluation

5 tracked models have a published HealthBench (consensus) score. Higher is better. Scores come from published lab reports and benchmark sources; recorded sources appear beside the scores.

## Ranking

| Rank | Model | Developer | Score | Source | Released |
| --- | --- | --- | --- | --- | --- |
| 1 | GPT-6 Sol | OpenAI | 95.8% | Lab | Sep 22 2026 |
| 2 | GPT-6 Luna | OpenAI | 95.6% | Lab | Sep 22 2026 |
| 3 | GPT-6 Astra | OpenAI | 95.4% | Lab | Sep 3 2026 |
| 4 | gpt-oss-120b | OpenAI | 89.9% | Lab | Aug 5 2025 |
| 5 | gpt-oss-20b | OpenAI | 82.6% | Lab | Aug 5 2025 |


---

Canonical page: https://aireleasetracker.com/benchmark/healthbench-consensus
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.