HELM SeaHELM - UIT-VSFC — leaderboard

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

Top models

#ModelScore
1Llama 3.1 70B Instruct73.08
2Llama 3.3 70B Instruct70.73
3Qwen 2 72B Instruct70.02
4GPT-4o Mini (2024-07-18)69.12
5Gemini 2.0 Flash68.87
6Llama 3.1 405B Instruct66.42
7DeepSeek V365.82
8GPT-4o (2024-11-20)65.67
9Gemini 2.0 Flash Lite65.61
10Llama 4 Scout Instruct65.49
11Llama 4 Maverick Instruct FP865.34
12Qwen 2.5 72B Instruct64.18
13Claude 3.7 Sonnet (20250219)61.71
14Llama 3.2 90B Vision Instruct33.59
15DeepSeek R1 Distill Llama 8B5.44

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