UbuntuGuard - English: leaderboard

Metric: F1 (%) for PASS/FAIL policy-compliance classification of user-agent dialogues, English policies and dialogues. Source: arxiv.org. Saturation forecast: Estimated already saturated. 15 models tracked.

Top models

#ModelScore
1Llama 3.3 70B Instruct98.51
2GPT-OSS-20B98.48
3DeepSeek V3.198.42
4gpt-oss-safeguard-20B97.26
5GPT-OSS-120B94.24

Interactive version: theaggregate.ai/benchmark?slug=ubuntuguard-english · How It Works · Data refreshed daily, snapshot 2026-09-26.