HELM SeaHELM - MLHSD — leaderboard
Metric: Macro F1 score. Source: crfm.stanford.edu. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.7 Sonnet (20250219) | 73.81 |
| 2 | GPT-4o (2024-11-20) | 72.31 |
| 3 | Llama 4 Maverick Instruct FP8 | 68.92 |
| 4 | Gemini 2.0 Flash | 68.55 |
| 5 | Llama 3.1 405B Instruct | 67.96 |
| 6 | Llama 3.3 70B Instruct | 66.39 |
| 7 | Qwen 2 72B Instruct | 66.01 |
| 8 | Llama 3.1 70B Instruct | 65.32 |
| 9 | Llama 4 Scout Instruct | 64.59 |
| 10 | GPT-4o Mini (2024-07-18) | 64.44 |
| 11 | Qwen 2.5 72B Instruct | 63.02 |
| 12 | Gemini 2.0 Flash Lite | 61.75 |
| 13 | Gemma 2 27B (IT) | 43.56 |
| 14 | Gemma 2 9B (IT) | 27.54 |
| 15 | DeepSeek V3 | 23.54 |
Interactive version: theaggregate.ai/benchmark?slug=helm-seahelm-mlhsd · How the rankings work · Data refreshed daily, snapshot 2026-07-22.