Google Carbon Report - HealthBench Pro, without Search grounding: leaderboard
Google DeepMind Carbon safety report, Table 5.1 / Figure 5.3. HealthBench Pro expert/clinician scenarios; native primary rubric score; WITHOUT Search grounding. Metric: Score (%). Higher is better. Source: https://storage.googleapis.com/deepmind-media/gemini/gemini_4_carbon_safety_report.pdf#page=61
Metric: Score (%). Source: storage.googleapis.com. Saturation forecast: Around January 2028. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Carbon | 72.8 | #2 |
| 2 | Gemini 4 Argon | 68.8 | #11 |
| 3 | Gemini 3.8 Flash | 65.4 | #36 |
No result here: #1 Claude Fable 5.2, #5 Claude Opus 5.5, #6 GPT-6 Astra.
Interactive version: theaggregate.ai/benchmark?slug=google-carbon-report-healthbench-pro-without-search-grounding · How It Works · Data refreshed daily, snapshot 2026-11-12.