RABBITS: leaderboard
DrugMatchQA accuracy for matching brand and generic drug names, part of the RABBITS biomedical drug-name robustness study.
Metric: Accuracy (%). Source: huggingface.co. 23 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 | 99.71 | #438 |
| 2 | GPT-4o | 99.61 | #343 |
| 3 | GPT-3.5 Turbo | 99.43 | #865 |
| 4 | Claude 3 Opus | 99.14 | #571 |
| 5 | Qwen 2 72B | 99.05 | #521 |
| 6 | Mixtral-8x22B-v0.1 | 98.66 | #711 |
| 7 | Llama 3 70B | 98.57 | #631 |
| 8 | c4ai-command-r-plus | 98.28 | #724 |
| 9 | Yi 1.5 34B | 97.81 | #807 |
| 10 | Mixtral 8x7B (v0.1) | 97.14 | #1083 |
| 11 | Llama 3 8B | 96.18 | #1242 |
| 12 | aya-23-35B | 95.81 | #883 |
| 13 | Llama 2 70B Base | 93.61 | #1025 |
| 14 | Qwen 2 7B | 93.13 | #1095 |
| 15 | Mistral-7B-v0.3 | 90.65 | #1355 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=rabbits · How It Works · Data refreshed daily, snapshot 2026-10-20.