AA-Omniscience v1.1 Hallucination Rate: leaderboard
Artificial Analysis's AA-Omniscience v1.1 factual-recall benchmark (3,000 questions across six domains): of the questions a model did not answer correctly, the share it answered incorrectly instead of abstaining. Reported 0-100 (%); lower is better. Declining to answer lowers the rate, so near-silent small models score best; read it beside AA-Omniscience v1.1 Accuracy. Scores are not directly comparable with v1, which had 6,000 questions.
Metric: Hallucination Rate (%). Source: artificialanalysis.ai. Saturation forecast: Not forecast. 134 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Carbon (Medium) | 4.24 | #2 (Gemini 4 Carbon) |
| 2 | Gemini 4 Argon (High) | 4.26 | #11 (Gemini 4 Argon) |
| 3 | Gemini 4 Carbon (High) | 11.42 | #2 (Gemini 4 Carbon) |
| 4 | Command A+ | 13.86 | #907 |
| 5 | Muse Ember (xHigh) | 17.34 | #54 (Muse Ember) |
| 6 | Muse Ember (Max) | 17.36 | #54 (Muse Ember) |
| 7 | Muse Wonder Full (Max) | 17.56 | #16 (Muse Wonder Full) |
| 8 | Qwen 3.8 27B (Non-reasoning) | 18.01 | #199 (Qwen 3.8 27B) |
| 9 | Nanbeige4.3-33B | 18.19 | #384 |
| 10 | Gemini 4 Carbon (Low) | 19.76 | #2 (Gemini 4 Carbon) |
| 11 | Grok 4.7 (xHigh) | 20.01 | #43 (Grok 4.7) |
| 12 | Grok 4.7 (High) | 20.21 | #43 (Grok 4.7) |
| 13 | Muse Wonder (Max) | 20.87 | #27 (Muse Wonder) |
| 14 | Solar Pro 4 | 20.92 | #456 |
| 15 | Muse Wonder (High) | 22.17 | #27 (Muse Wonder) |
No result here: #1 Claude Fable 5.2, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=aa-omniscience-v1-1-hallucination-rate · How It Works · Data refreshed daily, snapshot 2026-11-12.