HELM SeaHELM - Wisesight — leaderboard

Metric: Macro F1 score. Source: crfm.stanford.edu. 21 models tracked.

Top models

#ModelScore
1Qwen 2.5 72B Instruct55.49
2Claude 3.7 Sonnet (20250219)55
3GPT-4o (2024-11-20)54.76
4Gemini 2.0 Flash Lite54.73
5Gemini 2.0 Flash54.52
6Qwen 2 72B Instruct53.54
7GPT-4o Mini (2024-07-18)52.29
8DeepSeek V350.7
9Llama 4 Maverick Instruct FP849.7
10Claude 3.5 Haiku (20241022)48.25
11Llama 4 Scout Instruct46.58
12Llama 3.1 405B Instruct43.96
13Llama 3.1 70B Instruct39.63
14Llama 3.3 70B Instruct38.76
15Llama 3.2 90B Vision Instruct33.43

Interactive version: theaggregate.ai/benchmark?slug=helm-seahelm-wisesight · How the rankings work · Data refreshed daily, snapshot 2026-07-22.