Vectara Hallucination Leaderboard
Vectara Hallucination Leaderboard is run by read by the crawler. It has ranked 105 models, of which the catalogue holds 81, scoring from 3.1 to 24.2 on Hallucination Rate.
measured by read by the crawler · the board itself
| Place | Model | Metric | Score |
|---|---|---|---|
| 2ndof 105 | GPT-5.4 nano | Hallucination Rate | 3.1 |
| 3rdof 105 | Gemini 2.5 Flash-Lite | Hallucination Rate | 3.3 |
| 4thof 105 | Phi-4 | Hallucination Rate | 3.7 |
| 5thof 105 | Llama-3.3-70B-Instruct-Turbo | Hallucination Rate | 4.1 |
| 7thof 105 | Gemma 3 12B | Hallucination Rate | 4.4 |
| 9thof 105 | Qwen3 8B | Hallucination Rate | 4.8 |
| 10thof 105 | Nova Pro | Hallucination Rate | 5.1 |
| 11thof 105 | Nova 2 Lite | Hallucination Rate | 5.1 |
| 13thof 105 | Granite 4.0 H Small | Hallucination Rate | 5.2 |
| 16thof 105 | DeepSeek V3.2 Exp | Hallucination Rate | 5.3 |
| 17thof 105 | Qwen3 14B | Hallucination Rate | 5.4 |
| 18thof 105 | Nova Micro | Hallucination Rate | 5.5 |
| 19thof 105 | DeepSeek-V3.1 | Hallucination Rate | 5.5 |
| 20thof 105 | GPT-5.4 mini | Hallucination Rate | 5.5 |
| 21stof 105 | GPT-4.1 fine-tuned | Hallucination Rate | 5.6 |
| 22ndof 105 | Qwen3 4B | Hallucination Rate | 5.7 |
| 23rdof 105 | Grok 3 | Hallucination Rate | 5.8 |
| 24thof 105 | Qwen3 32B | Hallucination Rate | 5.9 |
| 25thof 105 | Nova Lite | Hallucination Rate | 6.1 |
| 26thof 105 | DeepSeek-V3 | Hallucination Rate | 6.1 |
| 27thof 105 | DeepSeek v3.2 | Hallucination Rate | 6.3 |
| 28thof 105 | Gemma 3 4B | Hallucination Rate | 6.4 |
| 29thof 105 | Command R+ (08-2024) | Hallucination Rate | 6.9 |
| 31stof 105 | GPT-5.4 | Hallucination Rate | 7 |
| 32ndof 105 | Gemini 2.5 Pro | Hallucination Rate | 7 |
| 34thof 105 | Gemma 3 27B | Hallucination Rate | 7.4 |
| 35thof 105 | Gemma 4 31B | Hallucination Rate | 7.4 |
| 38thof 105 | Gemini 2.5 Flash | Hallucination Rate | 7.8 |
| 39thof 105 | Llama 4 Maverick 17B | Hallucination Rate | 8.2 |
| 40thof 105 | Gemini 3.1 Flash-Lite | Hallucination Rate | 8.2 |
| 41stof 105 | GPT-5.4 Pro | Hallucination Rate | 8.3 |
| 42ndof 105 | GPT-5.2 | Hallucination Rate | 8.4 |
| 43rdof 105 | DeepSeek V4 Pro | Hallucination Rate | 8.6 |
| 44thof 105 | MiniMax M2.5 | Hallucination Rate | 9.1 |
| 45thof 105 | Command A 03-2025 | Hallucination Rate | 9.3 |
| 46thof 105 | GPT-5.5 | Hallucination Rate | 9.3 |
| 48thof 105 | Qwen3 235B A22B | Hallucination Rate | 9.3 |
| 49thof 105 | Qwen3 Next 80B A3B Thinking | Hallucination Rate | 9.3 |
| 50thof 105 | GLM-4.7 Flash | Hallucination Rate | 9.3 |
| 52ndof 105 | GLM-4.6 | Hallucination Rate | 9.5 |
| 53rdof 105 | Nemotron 3 Nano 30B A3B | Hallucination Rate | 9.6 |
| 54thof 105 | GPT-4o (2024-08-06) | Hallucination Rate | 9.6 |
| 56thof 105 | Claude Haiku 4.5 | Hallucination Rate | 9.8 |
| 57thof 105 | GLM-5 | Hallucination Rate | 10.1 |
| 58thof 105 | Claude Sonnet 4 | Hallucination Rate | 10.3 |
| 59thof 105 | Gemini 3.1 Pro Preview | Hallucination Rate | 10.4 |
| 60thof 105 | Qwen3.5-Flash | Hallucination Rate | 10.5 |
| 61stof 105 | Qwen3.5-35B-A3B | Hallucination Rate | 10.5 |
| 62ndof 105 | GPT-5 nano | Hallucination Rate | 10.5 |
| 63rdof 105 | Claude Sonnet 4.6 | Hallucination Rate | 10.6 |
| 65thof 105 | Qwen3.5 Plus 2026-02-15 | Hallucination Rate | 10.7 |
| 67thof 105 | Kimi K2.6 | Hallucination Rate | 10.8 |
| 68thof 105 | Claude Opus 4.5 | Hallucination Rate | 10.9 |
| 69thof 105 | Aya Expanse 32B | Hallucination Rate | 10.9 |
| 70thof 105 | GPT-5.1 | Hallucination Rate | 10.9 |
| 71stof 105 | Qwen3.5-122B-A10B | Hallucination Rate | 11.2 |
| 72ndof 105 | DeepSeek-R1 | Hallucination Rate | 11.3 |
| 73rdof 105 | GLM-4.7 | Hallucination Rate | 11.7 |
| 74thof 105 | Claude Opus 4.1 | Hallucination Rate | 11.8 |
| 75thof 105 | MiniMax M2.1 | Hallucination Rate | 11.8 |
| 76thof 105 | Claude Opus 4 | Hallucination Rate | 12 |
| 77thof 105 | Claude Sonnet 4.5 | Hallucination Rate | 12 |
| 78thof 105 | Claude Opus 4.7 | Hallucination Rate | 12 |
| 79thof 105 | Qwen3.5-27B | Hallucination Rate | 12.1 |
| 81stof 105 | Claude Opus 4.6 | Hallucination Rate | 12.2 |
| 82ndof 105 | Mercury 2 | Hallucination Rate | 12.3 |
| 83rdof 105 | MiniMax M2.7 | Hallucination Rate | 12.9 |
| 84thof 105 | GPT-5 mini | Hallucination Rate | 12.9 |
| 85thof 105 | Gemini 3 Flash Preview | Hallucination Rate | 13.5 |
| 86thof 105 | gemini-3-pro-preview | Hallucination Rate | 13.6 |
| 87thof 105 | Kimi K2.5 | Hallucination Rate | 14.2 |
| 88thof 105 | GPT OSS 120B | Hallucination Rate | 14.2 |
| 91stof 105 | GPT-5 | Hallucination Rate | 14.7 |
| 93rdof 105 | Grok 4.1 Fast Non-Reasoning | Hallucination Rate | 17.8 |
| 95thof 105 | o4-mini | Hallucination Rate | 18.6 |
| 97thof 105 | Grok 4.1 Fast Reasoning | Hallucination Rate | 19.2 |
| 98thof 105 | Ministral 3 14B 2512 | Hallucination Rate | 19.4 |
| 99thof 105 | grok-4-fast-non-reasoning | Hallucination Rate | 19.7 |
| 101stof 105 | Ministral 3 8B 2512 | Hallucination Rate | 21.7 |
| 103rdof 105 | o3-pro | Hallucination Rate | 23.3 |
| 105thof 105 | Ministral 3 3B 2512 | Hallucination Rate | 24.2 |
Read from the board on 2026-09-02