Evaluation sources/BullshitBench v2
ParticipatingKnowledge & language
BullshitBench v2
BullshitBench · Public model evaluation results from BullshitBench v2.
Model results from this source
| Qwen3.8 Max | 0.95 | 80.1 | 2026-11-05 |
| Claude Opus 4.8 | 0.95 | 80.1 | 2026-11-05 |
| Claude Sonnet 4.6 | 0.91 | 77.8 | 2026-11-05 |
| Claude Opus 4.5 | 0.9 | 77.2 | 2026-11-05 |
| Claude Opus 4.6 | 0.87 | 75.4 | 2026-11-05 |
| Claude Opus 4.7 | 0.83 | 73.1 | 2026-11-05 |
| Claude Sonnet 5 | 0.8 | 71.4 | 2026-11-05 |
| Claude Sonnet 4.5 | 0.79 | 70.8 | 2026-11-05 |
| Qwen3.5 397B-A17B | 0.78 | 70.2 | 2026-11-05 |
| Claude Fable 5.1 | 0.77 | 69.7 | 2026-11-05 |
| Claude Haiku 4.5 | 0.77 | 69.7 | 2026-11-05 |
| Kimi K3 | 0.74 | 67.9 | 2026-11-05 |
| Claude Opus 5 | 0.73 | 67.3 | 2026-11-05 |
| Qwen3.6 Plus | 0.72 | 66.8 | 2026-11-05 |
| Qwen3.7 Max | 0.72 | 66.8 | 2026-11-05 |
| GLM 5.3 | 0.72 | 66.8 | 2026-11-05 |
| GPT-6 Astra | 0.69 | 65.0 | 2026-11-05 |
| Gemini 3.5 Flash Lite | 0.66 | 63.3 | 2026-11-05 |
| Gemini 4 Argon | 0.66 | 63.3 | 2026-10-18 |
| Gemini 4 Argon | 0.66 | 63.3 | 2026-11-05 |
| Gemini 4 Argon | 0.66 | 63.3 | 2026-10-17 |
| Grok 4.6 | 0.66 | 63.3 | 2026-11-05 |
| Kimi K2.6 | 0.65 | 62.7 | 2026-11-05 |
| GPT-6.1 Sol | 0.65 | 62.7 | 2026-11-05 |
Page 1 of 4