HELM MedHELM - Administration & Workflow: leaderboard
HELM MedHELM administration-and-workflow generation category, scored by mean LLM-jury win rate.
Metric: Mean win rate. Source: crfm.stanford.edu. Saturation forecast: Around January 2027. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 86.67 | #98 |
| 2 | GPT-5 Mini (2025-08-07) | 68.33 | #172 |
| 3 | Claude 3.5 Sonnet (20241022) | 66.67 | #291 |
| 4 | GPT-4o (2024-05-13) | 60.83 | #325 |
| 5 | Claude 3.7 Sonnet (20250219) | 58.33 | #201 |
| 6 | GPT-4o Mini (2024-07-18) | 55.83 | #558 |
| 7 | O4 Mini (2025-04-16) | 54.17 | #179 |
| 8 | DeepSeek R1 | 47.5 | #251 |
| 9 | O3 Mini (2025-01-31) | 40.83 | #240 |
| 10 | Gemini 2.5 Pro (Preview 05-06) | 34.17 | #147 |
| 11 | Gemini 1.5 Pro (001) | 29.17 | #444 |
| 12 | Gemini 2.0 Flash | 24.17 | #342 |
| 13 | Llama 3.3 70B Instruct | 23.33 | #534 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=helm-medhelm-administration-workflow · How It Works · Data refreshed daily, snapshot 2026-10-20.