Factuality Evaluation on Bio
14.1PrecisionGemini2.5-flash-lite
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Gemini2.5-flash-lite2026.04 | 14.1 | 10.9 | 9.4 | |
| Llama-3.1-70B-Inst2026.04 | 14 | 9.9 | 8.3 | |
| GPT4o-mini2026.04 | 13.3 | 9.1 | 8.2 | |
| Mistral-7B-instruct-v0.32026.04 | 9.9 | 4.8 | 4.8 | |
| Llama-3.1-8B-Inst2026.04 | 8 | 5 | 4 | |
| Qwen2.5-7B-Instruct2026.04 | 6.3 | 3 | 3 |