HELM Lite - LegalBench - Corporate Lobbying: leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 91 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.1 70B Instruct | 87.35 |
| 2 | Gemini 1.5 Pro (001) | 86.12 |
| 3 | Llama 3.3 70B Instruct | 85.71 |
| 4 | GPT-4o Mini (2024-07-18) | 85.51 |
| 5 | Llama 3.2 90B Vision Instruct | 85.51 |
| 6 | Llama 3.1 405B Instruct | 85.31 |
| 7 | Qwen 2.5 72B Instruct | 85.1 |
| 8 | Gemma 2 9B (IT) | 84.49 |
| 9 | Gemini 1.5 Pro (002) | 84.29 |
| 10 | GPT-4o (2024-08-06) | 84.29 |
| 11 | Command-R+ | 84.29 |
| 12 | DeepSeek V3 | 84.08 |
| 13 | Mistral Large 2 (Jul) | 84.08 |
| 14 | Claude 3.5 Haiku (20241022) | 83.47 |
| 15 | Nova Pro | 83.47 |
Interactive version: theaggregate.ai/benchmark?slug=helm-lite-legalbench-corporate-lobbying · How It Works · Data refreshed daily, snapshot 2026-09-19.