Multi-modal Understanding on MMBench EN 1.1 (dev)
84.31AccuracyPixtral-12B
Evaluation results
| Method | Links | |
|---|---|---|
| Pixtral-12BParameters=12B, Training=Full fine-tuning2026.02 | 84.31 | |
| Qwen2.5-VL-7BParameters=7B2026.02 | 80.62 | |
| PaliGemma2-10BParameters=10B2026.02 | 80.46 | |
| LLaVA-PLLuM-12b-ncParameters=12B, Training=Parameter-efficient LoRA adaptation2026.02 | 78.43 | |
| LLaVA-Bielik-11b-v2.6Parameters=11B, Version=v2.6, Training=Parameter-efficient LoRA adaptation2026.02 | 77.75 | |
| LLaVA-1.6-Mistral-7BBackbone=Mistral-7B2026.02 | 76.54 | |
| LLaVA-PLLuM-12b-nc-250715Parameters=12B, Training=Parameter-efficient LoRA adaptation2026.02 | 75.23 | |
| LLaVA-1.6-Vicuna-13BBackbone=Vicuna-13B2026.02 | 74.39 |