List your company
ResearchFormerly WizWand

Visual Understanding on MM-Vet

76.9MM-Vet ScoreGPT-4o
35.19646.02356.8567.677Sep 7, 2024Dec 24, 2024Apr 12, 2025Jul 30, 2025Nov 16, 2025Mar 5, 2026Jun 22, 2026
Updated 3mo ago

Evaluation results

MethodLinks
GPT-4o2025.0676.9
LVRPOType=Unified (>1.5B), # LLM Params=7B2026.0369.5
GPT-4o-05132024.0969.1
BAGEL2025.0667.2
BAGELType=Unified, LLM Parameters=7B MOT2025.0567.2
BAGELType=Unified (>1.5B), # LLM Params=7B2026.0367.2
BAGELModel Category=Unified Understanding & Generation Model2025.1067.2
UniWorld-V12025.0667.1
Qwen2.5-VLType=Und. Only, LLM Parameters=7B2025.0567.1
Qwen2.5-VLType=Und. Only, # LLM Params=7B2026.0367.1
Kimi-VLType=Und. Only, LLM Parameters=2.8B/16B2025.0566.7
Kimi-VLType=Und. Only, # LLM Params=2.8B/16B2026.0366.7
BLIP3-o2025.0666.6
MetaQuery2025.0666.6
MetaQuery-XLType=Unified, LLM Parameters=7B2025.0566.6
MetaQuery-XLType=Unified (>1.5B), # LLM Params=7B2026.0366.6
UniVideoModel Category=Unified Understanding & Generation Model, Backbone=Qwen-2.5VL-7B2025.1066.6
Claude3.5-Sonnet2024.0966
Gemini-1.5-Pro2024.0964
InternVL2.5Type=Und. Only, LLM Parameters=7B2025.0562.8
InternVL2.5Type=Und. Only, # LLM Params=7B2026.0362.8
InternVL3-2BLLM Params=2B, Model Category=Und. Only2026.0662.2
SPAR-3BLLM Params=2B, Model Category=Und. and Gen.2026.0662.2
Qwen2-VLType=Und. Only, LLM Parameters=7B2025.0562
Qwen2-VLType=Und. Only, # LLM Params=7B2026.0362
Qwen2.5-VLType=Und. Only, LLM Parameters=3B2025.0561.8
Qwen2.5-VLType=Und. Only, # LLM Params=3B2026.0361.8
Qwen2.5-VL-3BLLM Params=3B, Model Category=Und. Only2026.0661.8
OmniGen2Model Category=Unified Understanding & Generation Model2025.1061.8
InternVL2.5Type=Und. Only, LLM Parameters=1.8B2025.0560.8
InternVL2.5Type=Und. Only, # LLM Params=1.8B2026.0360.8
BLIP3-o-4BLLM Params=4B, Model Category=Und. and Gen.2026.0660.1
DeepSeek-VL2Type=Und. Only, LLM Parameters=4.1B/28B2025.0560
DeepSeek-VL2Type=Und. Only, # LLM Params=4.1B/28B2026.0360
SPAR-1BLLM Params=1B, Model Category=Und. and Gen.2026.0659.8
InternVL3-1BLLM Params=1B, Model Category=Und. Only2026.0659.5
LLava-OVType=Und. Only, LLM Parameters=7B2025.0557.5
LLava-OVType=Und. Only, # LLM Params=7B2026.0357.5
LLaVA-NeXTModel Category=Video Understanding Model2025.1057.4
Show-o2Model Category=Unified Understanding & Generation Model2025.1056.6
DPA-Qwen3-32BTarget LLM=Qwen3-32B2026.0556
MUSE-VLType=Unified, LLM Parameters=32B2025.0555.9
InternVL2-8B2024.0954.3
InternVL2Type=Und. Only, LLM Parameters=7B2025.0554.2
InternVL2Type=Und. Only, # LLM Params=7B2026.0354.2
Cambrian-34B2024.0953.2
LVRPOType=Unified, # LLM Params=1.5B2026.0353.2
POINTS-7BLLM=Qwen-2.5-7B2024.0952.3
OneVision2024.0951.9
Ovis1.5-LLAMA3-8B2024.0950.9
Janus-ProType=Unified, LLM Parameters=7B2025.0550
POINTS-9BLLM=Yi-1.5-9B2024.0950
Janus-ProType=Unified (>1.5B), # LLM Params=7B2026.0350
Janus-Pro-7BLLM Params=7B, Model Category=Und. and Gen.2026.0650
Qwen2-VLType=Und. Only, LLM Parameters=1.5B2025.0549.5
Qwen2-VLType=Und. Only, # LLM Params=1.5B2026.0349.5
Qwen2-VL-2BTarget LLM=Public Models2026.0549.5
IXC-2.52024.0949.3
LLaVA-NeXT-Qwen3-32BTarget LLM=Qwen3-32B2026.0549.2
EMU-2Type=Und. & Gen. Non-prob., LLM=LLaMA-13B, V-Token=CLIP, Resolution=4482024.1048.5
BAGELType=Unified, LLM Parameters=1.5B MOT2025.0548.2
BAGELType=Unified, # LLM Params=1.5B2026.0348.2
BAGEL-7BLLM Params=3B, Model Category=Und. and Gen.2026.0648.2
TokenFlow-XLModel Category=Unified Understanding & Generation Model2025.1048.2
InternVL2Type=Und. Only, LLM Parameters=1.8B2025.0544.6
InternVL2Type=Und. Only, # LLM Params=1.8B2026.0344.6
Cambrian-1-8BTarget LLM=Public Models2026.0544.2
SEED-XType=Unified, LLM Parameters=13B2025.0543
SEED-XType=Unified (>1.5B), # LLM Params=13B2026.0343
SEED-X-13BLLM Params=13B, Model Category=Und. and Gen.2026.0643
DPA-Qwen3-4BTarget LLM=Qwen3-4B2026.0542.7
DPA-LLaMA-3.2-3BTarget LLM=LLaMA-3.2-3B2026.0542
Idefics3-LLAMA3-8B2024.0941.7
LLaVA-NeXT-Qwen3-4BTarget LLM=Qwen3-4B2026.0541.4
LLaVA1.5-13B-BPOModel Size=13B, Baseline Category=reinforcement learning-based2024.1141.4
TokenFlow-XLType=Unified, LLM Parameters=13B2025.0540.7
TokenFlow-XLType=Unified (>1.5B), # LLM Params=13B2026.0340.7
DualTokenLLM=LLaMA-2-7B, Resolution=2562026.0340.5
EvoTokLLM=Qwen-2.5-7B-Inst., Resolution=2562026.0339.9
LACINGModel Size=13B2024.1139.9
Janus-Pro2025.0639.8
Janus-ProType=Unified, LLM Parameters=1.5B2025.0539.8
Janus-ProType=Unified, # LLM Params=1.5B2026.0339.8
LLaVA-NeXT-7BTarget LLM=Public Models2026.0539
ShareGPT4VLLM=Vicuna-7B, Resolution=3362026.0337.6
Dynamic-LLaVA-13BILLM=Vicuna-13B, Res.=336, Token=115 (-80%), TFLOPS=4.7+0.01 (-76%)2024.1237.3
Emu32025.0637.2
Emu3-ChatType=Und. Only, LLM Parameters=8B2025.0537.2
EMU-3Type=Joint Prob. Models, LLM=8B from scratch, V-Token=vq-vae, Resolution=5122024.1037.2
EMU3LLM=8B from scratch, Resolution=-2026.0337.2
Emu3-ChatType=Und. Only, # LLM Params=8B2026.0337.2
Emu3-Chat-8BLLM Params=8B, Model Category=Und. Only2026.0637.2
Emu3Model Category=Unified Understanding & Generation Model2025.1037.2
Dynamic-LLaVA-TokenPacker-13B_VTProjector=TokenPacker, Image Resolution=336, Prefill Tokens=57, Prefill TFLOPs=2.12024.1237.1
VIG training + LACINGBase Model=LLaVA-1.5 7B2026.0237.01
ILLUMEType=Unified, LLM Parameters=7B2025.0537
ILLUMELLM=Vicuna-7B, Resolution=-2026.0337
ILLUMEType=Unified (>1.5B), # LLM Params=7B2026.0337
VCDModel Size=13B, Baseline Category=training-free2024.1136.9
LLaVA1.5-7B-BPOModel Size=7B, Baseline Category=reinforcement learning-based2024.1136.8