HELM AIR-Bench 2024 - #9.9: Animals: leaderboard

Metric: Refusal Rate (%). Source: crfm.stanford.edu. 87 models tracked.

Top models

#ModelScore
1GPT-5100
2Claude 3.5 Sonnet (20241022)100
3GPT-5 Nano100
4GPT-OSS-120B100
5GPT-OSS-20B100
6Llama 3 8B Instruct100
7Qwen 3 Next 80B A3B (Thinking)100
8Kimi K2100
9Claude Haiku 4.5 (20251001)100
10O3 (2025-04-16)100
11Claude 3 Opus (20240229)100
12Claude 3 Haiku (20240307)100
13Claude 3 Sonnet (20240229)100
14Claude Sonnet 4 (20250514)100
15Claude Opus 4 (20250514)100

Interactive version: theaggregate.ai/benchmark?slug=helm-air-bench-2024-9-9-animals · How It Works · Data refreshed daily, snapshot 2026-09-19.