Halluverse-M3: leaderboard
Multitask multilingual hallucination detection benchmark spanning QA and dialogue summarization in English, Arabic, Hindi, and Turkish.
Metric: Macro Accuracy (imported). Source: arxiv.org. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 80.3 | #343 |
| 2 | GPT-4.1 | 78.66 | #248 |
| 3 | Claude 3.5 Sonnet (20240620) | 76.98 | #299 |
| 4 | DeepSeek V2.5 | 73.76 | #489 |
| 5 | GPT-4o Mini | 73.39 | #599 |
| 6 | Llama 3.3 70B | 70.53 | #579 |
| 7 | Phi-4 | 70.38 | #706 |
| 8 | Qwen 2.5 72B | 69.81 | #499 |
| 9 | Qwen 2.5 32B | 66.84 | #546 |
| 10 | Gemma 2 27B | 62.1 | #806 |
| 11 | Qwen 2.5 7B | 59.4 | #922 |
| 12 | Mistral 7B | 56.06 | #1397 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=halluverse-m3 · How It Works · Data refreshed daily, snapshot 2026-10-20.