Multimodal Understanding on MME (test)
2,473.6ScoreMiMo-VL-7B-SFT-2508
Evaluation Results
| Method | Links | |
|---|---|---|
| MiMo-VL-7B-SFT-2508Evaluation pipeline=reproduced2025.12 | 2,473.6 | |
| MiMo-VL-Miloco-7BEvaluation pipeline=reproduced2025.12 | 2,396.4 | |
| Lumina-DiMOOParams=8B, Base model=LLaDA, Architecture=Discrete Diff., Data scale=80M2025.05 | 1,534.2 | |
| MMaDAParams=8B, Base model=LLaDA, Architecture=Discrete Diff., Data scale=Unknown, Resolution=512 × 5122025.05 | 1,410.7 | |
| MudditParams=1B, Base model=Meissonic, Architecture=Discrete Diff, Data scale=16M, Resolution=1024 × 10242025.05 | 1,139.2 | |
| D-DiTParams=2B, Base model=SD3-medium, Architecture=Discrete Diff. + Diff., Data scale=40M, Resolution=512 × 5122025.05 | 1,124.7 | |
| MudditParams=1B, Base model=Meissonic, Architecture=Discrete Diff., Data scale=10M, Resolution=512 × 5122025.05 | 1,107.4 | |
| Show-OParams=1.3B, Base model=Phi-1.5, Architecture=AR + Discrete Diff, Data scale=35M, Resolution=512 × 5122025.05 | 1,097.2 |