BBQ Disambiguated Accuracy (Mythos 5.1 System Card): leaderboard
Accuracy on disambiguated social-context questions in BBQ, without a system prompt. September 2026 evaluation infrastructure; earlier models were re-evaluated. Measures answering the supplied evidence correctly, not the signed bias statistic.
Metric: Accuracy (%). Source: www-cdn.anthropic.com. Saturation forecast: Around February 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Mythos 5.1 | 89.9 | #2 |
| 2 | Claude Mythos 5 | 84.1 | #8 |
| 3 | Claude Sonnet 5 (Non-reasoning) | 72.36 | #65 (Claude Sonnet 5) |
No result here: #6 GPT-6 Astra, #7 GPT-6.1 Sol, #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=bbq-disambiguated-accuracy-mythos-5-1-system-card · How It Works · Data refreshed daily, snapshot 2026-11-05.