Vision-Language Understanding on MM-Vet
72.16Total ScoreREVIS
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| REVIS2026.02 | 72.16 | 67.93 | 78.85 | 61.55 | 61.37 | 72.53 | 84.62 | — | |
| Regulardecoding=standard greedy decoding baseline2026.02 | 70.18 | 65.67 | 76.67 | 58.57 | 59 | 70.4 | 72.69 | — | |
| Only2026.02 | 70.05 | 64.87 | 77.92 | 56.9 | 59.5 | 71.6 | 76.54 | — | |
| AGLA2026.02 | 69.56 | 65.03 | 76.77 | 58.63 | 60.38 | 68.4 | 72.69 | — | |
| M3ID2026.02 | 69.43 | 64.83 | 76.77 | 58.99 | 61.5 | 71.07 | 76.54 | — | |
| VCD2026.02 | 68.76 | 64.8 | 73.96 | 57.74 | 60.12 | 66.27 | 72.69 | — | |
| VTI2026.02 | 56.38 | 48.53 | 66.87 | 39.88 | 37.75 | 63.07 | 76.92 | — | |
| Qwen-VL-Chat2024.05 | 47.3 | — | — | — | — | — | — | — | |
| VILA-7BSIMA=true2024.05 | 38.4 | — | — | — | — | — | — | — | |
| Emu-I-14Bzero-shot=true, parameter_size=14B2023.07 | 36.3 | 45.5 | 19.2 | 36.7 | 35.9 | 25.2 | 3.8 | — | |
| mPLUG-Owl22024.05 | 36.2 | — | — | — | — | — | — | — | |
| SENTINELModel=LLaVA-v1.5-13B2026.02 | 36.2 | — | — | — | — | — | — | — | |
| HII-DPOModel=LLaVA-v1.5-13B2026.02 | 36.2 | — | — | — | — | — | — | — | |
| baselineModel=LLaVA-v1.5-13B2026.02 | 36 | — | — | — | — | — | — | — | |
| DreamLLM-7Bzero-shot=true, parameter_size=7B2023.07 | 35.9 | 41.8 | 26.4 | 33.4 | 33 | 31 | 11.5 | — | |
| LLaVA-65Bzero-shot=true, parameter_size=65B2023.07 | 35.5 | 39.2 | 28.2 | 26.2 | 28.3 | 33 | 15 | — | |
| vanilla-DPOModel=LLaVA-v1.5-13B2026.02 | 35 | — | — | — | — | — | — | — | |
| VILA-7BSIMA=false2024.05 | 34.9 | — | — | — | — | — | — | — | |
| HSA-DPOModel=LLaVA-v1.5-13B2026.02 | 34 | — | — | — | — | — | — | — | |
| HII-DPOModel=LLaVA-v1.5-7B2026.02 | 33.8 | — | — | — | — | — | — | — | |
| Re-AlignModel=LLaVA-v1.5-13B2026.02 | 33.5 | — | — | — | — | — | — | — | |
| SENTINELModel=LLaVA-v1.5-7B2026.02 | 32.6 | — | — | — | — | — | — | — | |
| POVIDModel=LLaVA-v1.5-7B2026.02 | 31.8 | — | — | — | — | — | — | — | |
| LLaVA-1.5-7BSIMA=true2024.05 | 31.6 | — | — | — | — | — | — | — | |
| baselineModel=LLaVA-v1.5-7B2026.02 | 31 | — | — | — | — | — | — | — | |
| HA-DPOModel=LLaVA-v1.5-7B2026.02 | 30.6 | — | — | — | — | — | — | — | |
| LLaVA1.5Size=7B2024.01 | 30.5 | — | — | — | — | — | — | — | |
| LLaVA-1.5-7BSIMA=false2024.05 | 30.5 | — | — | — | — | — | — | — | |
| IDEFICS2024.05 | 30 | — | — | — | — | — | — | — | |
| RLAIF-VModel=LLaVA-v1.5-7B2026.02 | 29.9 | — | — | — | — | — | — | — | |
| Vary-toySize=1.8B2024.01 | 29 | 33.4 | 20.3 | 19.9 | 17.5 | 24 | 10.8 | — | |
| Re-AlignModel=LLaVA-v1.5-7B2026.02 | 28.8 | — | — | — | — | — | — | — | |
| InstructBLIP2024.05 | 26.2 | — | — | — | — | — | — | — | |
| TPOModel=LLaVA-v1.5-7B2026.02 | 25.7 | — | — | — | — | — | — | — | |
| InstructBLIP-14Bzero-shot=true, parameter_size=14B2023.07 | 25.6 | 30.8 | 16 | 9.8 | 9 | 21.1 | 10.5 | — | |
| OpenFlamingo2024.01 | 24.8 | 28.7 | 16.7 | 16.4 | 13.1 | 21 | 7.7 | — | |
| Otter2024.01 | 24.7 | 27.3 | 17.8 | 14.2 | 13.8 | 24.4 | 3.8 | — | |
| MiniGPT-42024.01 | 24.4 | 29.9 | 16.1 | 20.4 | 22.1 | 22.2 | 3.8 | — | |
| MiniGPT-4-14Bzero-shot=true, parameter_size=14B2023.07 | 24.4 | 29.9 | 16.1 | 20.4 | 22.1 | 22.2 | 3.8 | — | |
| LLaVASize=7B2024.01 | 23.8 | 28 | 17.1 | 16.3 | 18.9 | 21.2 | 11.5 | — | |
| BLIP-22024.01 | 22.4 | 27.5 | 11.1 | 11.8 | 7 | 16.2 | 5.8 | — | |
| BLIP-22024.05 | 22.4 | — | — | — | — | — | — | — | |
| LLaMA-Adapter v2-7Bzero-shot=true, parameter_size=7B2023.07 | 13.6 | 16.8 | 7.8 | 2.5 | 3 | 16.6 | 4.4 | — | |
| AdaMoENA=1.39B, Avgk=6.132026.06 | — | — | — | — | — | — | — | 40.12 | |
| AdashieldModel=llava-llama2-7b2025.05 | — | 36.2 | 30.9 | 32.7 | 34.4 | 31.2 | 18.5 | — | |
| BaseModel=llava-llama2-7b2025.05 | — | 50.3 | 33.4 | 43.5 | 45 | 36.8 | 29.2 | — | |
| CoCAModel=llava-llama2-7b2025.05 | — | 28.7 | 17.4 | 25 | 26.5 | 21.5 | 16.9 | — | |
| DTRModel=llava-llama2-7b2025.05 | — | 50.3 | 30.6 | 40.7 | 44.4 | 39.1 | 23.8 | — | |
| DYNMoENA=1.40B, Avgk=6.192026.06 | — | — | — | — | — | — | — | 38.9 | |
| GeMoENA=1.32B, Avgk=5.432026.06 | — | — | — | — | — | — | — | 44.1 | |
| GRPOModel Scale=0.8B, Prompt Replay Buffer Augmentation=false2026.06 | — | — | — | — | — | — | — | 55.6 | |
| GRPOModel Scale=0.8B, Prompt Replay Buffer Augmentation=true2026.06 | — | — | — | — | — | — | — | 57.5 | |
| GRPOModel Scale=2B, Prompt Replay Buffer Augmentation=false2026.06 | — | — | — | — | — | — | — | 72.3 | |
| GRPOModel Scale=2B, Prompt Replay Buffer Augmentation=true2026.06 | — | — | — | — | — | — | — | 74 | |
| MoE++NA=1.48B, Avgk=7.032026.06 | — | — | — | — | — | — | — | 38.4 | |
| MolmoE-1B-7BNA=1.58B, Avgk=8.002026.06 | — | — | — | — | — | — | — | 45.6 | |
| Off-DistillModel Scale=0.8B, Prompt Replay Buffer Augmentation=false2026.06 | — | — | — | — | — | — | — | 51.5 | |
| Off-DistillModel Scale=0.8B, Prompt Replay Buffer Augmentation=true2026.06 | — | — | — | — | — | — | — | 51.8 | |
| Off-DistillModel Scale=2B, Prompt Replay Buffer Augmentation=false2026.06 | — | — | — | — | — | — | — | 69.2 | |
| Off-DistillModel Scale=2B, Prompt Replay Buffer Augmentation=true2026.06 | — | — | — | — | — | — | — | 69.2 | |
| On-DistillModel Scale=0.8B, Prompt Replay Buffer Augmentation=false2026.06 | — | — | — | — | — | — | — | 52.7 | |
| On-DistillModel Scale=0.8B, Prompt Replay Buffer Augmentation=true2026.06 | — | — | — | — | — | — | — | 52.5 | |
| On-DistillModel Scale=2B, Prompt Replay Buffer Augmentation=false2026.06 | — | — | — | — | — | — | — | 69.2 | |
| On-DistillModel Scale=2B, Prompt Replay Buffer Augmentation=true2026.06 | — | — | — | — | — | — | — | 69.4 | |
| Qwen3.5Model Scale=0.8B, Prompt Replay Buffer Augmentation=false2026.06 | — | — | — | — | — | — | — | 53.2 | |
| Qwen3.5Model Scale=2B, Prompt Replay Buffer Augmentation=false2026.06 | — | — | — | — | — | — | — | 69.7 | |
| ShiftDCModel=llava-llama2-7b2025.05 | — | 44.7 | 31.5 | 40.2 | 44 | 33.6 | 23.4 | — | |
| Top-pNA=2.05B, Avgk=12.652026.06 | — | — | — | — | — | — | — | 43.1 | |
| ZPPOModel Scale=0.8B, Prompt Replay Buffer Augmentation=true2026.06 | — | — | — | — | — | — | — | 59.9 | |
| ZPPOModel Scale=2B, Prompt Replay Buffer Augmentation=true2026.06 | — | — | — | — | — | — | — | 77.1 |