EVALITA - hate-speech-detection — leaderboard
Metric: CPS. Source: huggingface.co. 48 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.3 70B Instruct | 77.82 |
| 2 | Mistral Large 2 (Nov) Instruct (2411) | 77.03 |
| 3 | Qwen 2.5 72B Instruct | 76.38 |
| 4 | Gemma 3 27B (IT) | 75.87 |
| 5 | calme-3.2-instruct-78B | 75.82 |
| 6 | Qwen 2.5 VL 32B Instruct | 75.47 |
| 7 | DeepSeek R1 Distill Llama 70B | 74.6 |
| 8 | Gemma 2 27B (IT) | 74.14 |
| 9 | Mistral Small 3 | 72.9 |
| 10 | Qwen2.5-14B-Instruct-1M | 72.8 |
| 11 | Gemma 2 9B (IT) | 71.99 |
| 12 | Gemma 3 12B (IT) | 71.4 |
| 13 | LLaMAntino-3-ANITA-8B-Inst-DPO-ITA | 70.99 |
| 14 | Qwen 3 Next 80B A3B Instruct | 69.84 |
| 15 | Llama-3.1-SuperNova-Lite | 69.24 |
Interactive version: theaggregate.ai/benchmark?slug=evalita-hate-speech-detection · How the rankings work · Data refreshed daily, snapshot 2026-07-22.