List your company
ResearchFormerly WizWand

General Task on MMStar (accuracy)

76.2AccuracySeed1.5-VL
30.85642.62854.466.172Oct 2, 2025Nov 11, 2025Dec 21, 2025Jan 30, 2026Mar 11, 2026Apr 20, 2026May 30, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Seed1.5-VLModel=Seed1.5-VL, Training Strategy=Reference2025.1076.2
Qwen2.5-VL-72B-InstructModel=Qwen2.5-VL-72B-Instruct, Training Strategy=Reference2025.1070.8
Qwen2.5-VL-32B-InstructModel=Qwen2.5-VL-32B-Instruct, Training Strategy=Reference2025.1069.5
InternVL3-8BModel Category=Open-source Models2026.0568.2
Thyme-7BModel Category=Tool-based Models2026.0565.9
DeepLatent-RL-7BModel Category=Latent Visual reasoning Models, Training Stage=RL, Training Data=full DeepLatent-180K dataset2026.0565
DeepLatent-RL-7B*Model Category=Latent Visual reasoning Models, Training Stage=RL, Training Data=visual search data of DeepLatent-180K in SFT Stage 22026.0564.1
DeepLatent-SFT-7BModel Category=Latent Visual reasoning Models, Training Stage=SFT2026.0563.6
AODModel=Qwen2.5-VL-7B2026.0563
Qwen2.5-VL-7B-Instruct + SFT -> RLModel=Qwen2.5-VL-7B-Instruct, Training Strategy=SFT -> RL Baseline2025.1062.27
REWARDMAPModel=Qwen2.5-VL-7B-Instruct, Training Strategy=REWARDMAP2025.1062.27
LLaVA-OneVision-7BModel Category=Open-source Models2026.0561.9
Qwen2.5-VL-7BModel Category=Open-source Models2026.0561.9
Kimi-VL-A3B-InstructModel=Kimi-VL-A3B-Instruct, Training Strategy=Reference2025.1061.7
Qwen2.5-VL-7B-InstructModel=Qwen2.5-VL-7B-Instruct, Training Strategy=Base Model2025.1061.67
TruthPrIntModel=Qwen2.5-VL-7B2026.0561.1
ASDModel=Qwen2.5-VL-7B2026.0560.7
AODModel=InternVL3-8B2026.0560.2
VASparseModel=Qwen2.5-VL-7B2026.0560.1
VCDModel=Qwen2.5-VL-7B2026.0559.6
PruneHalModel=InternVL3-8B2026.0559
BaseModel=Qwen2.5-VL-7B2026.0558.9
PruneHalModel=Qwen2.5-VL-7B2026.0557.8
TruthPrIntModel=InternVL3-8B2026.0556.8
ASDModel=InternVL3-8B2026.0556.1
VASparseModel=InternVL3-8B2026.0556
Qwen2.5-VL-3B-InstructModel=Qwen2.5-VL-3B-Instruct, Training Strategy=Reference2025.1055.9
BaseModel=InternVL3-8B2026.0555.8
VCDModel=InternVL3-8B2026.0554.6
AODModel=LLaVA-1.5-7B2026.0536.3
ASDModel=LLaVA-1.5-7B2026.0535.8
TruthPrIntModel=LLaVA-1.5-7B2026.0535.2
PruneHalModel=LLaVA-1.5-7B2026.0535.1
VCDModel=LLaVA-1.5-7B2026.0534.1
VASparseModel=LLaVA-1.5-7B2026.0533.5
BaseModel=LLaVA-1.5-7B2026.0532.6