ClinSQL - Disease Diagnosis and Outcomes: leaderboard
Metric: Execution Score (%). Source: arxiv.org. Saturation forecast: Around December 2026. 22 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 Chat | 80.22 |
| 2 | GPT-5 Mini | 79.55 |
| 3 | GPT-4.1 | 77.52 |
| 4 | Gemini 2.5 Pro | 76.43 |
| 5 | DeepSeek R1 0528 | 67.97 |
| 6 | Gemini 2.5 Flash | 67.32 |
| 7 | Qwen 3 Coder 480B A35B Instruct | 59.94 |
| 8 | Grok 4 Fast (Reasoning) | 58.75 |
| 9 | O4 Mini (2025-04-16) | 58.19 |
| 10 | DeepSeek V3.1 | 57.61 |
| 11 | Qwen 3 235B A22B 2507 Instruct | 55.79 |
| 12 | GPT-5 Nano | 54.28 |
| 13 | Llama 4 Maverick Instruct | 51.33 |
| 14 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 50.04 |
| 15 | Qwen 3 Next 80B A3B Instruct | 43.37 |
Interactive version: theaggregate.ai/benchmark?slug=clinsql-disease-diagnosis-and-outcomes · How It Works · Data refreshed daily, snapshot 2026-09-25.