List your company

Compositional Hallucination Evaluation on OmniVCHall Generated

94AccuracyHuman
47.259.3571.583.65Jan 31, 2026

Evaluation results

MethodLinks
HumanSize=-2026.0194
Qwen3-VL-ThinkingSize=235B2026.0178
Qwen3-VL-ThinkingSize=32B2026.0177
Qwen3-VL-InstructSize=235B2026.0177
GPT-5.2Size=-2026.0177
Gemini-3-proSize=-2026.0177
Doubao-Seed-1.8Size=-2026.0177
Qwen3-VL-InstructSize=32B2026.0176
GPT-5Size=-2026.0175
Gemini-2.5-flashSize=-2026.0175
Doubao-Seed-1.6Size=-2026.0175
Qwen3-VL-InstructSize=8B2026.0174
Qwen3-VL-ThinkingSize=8B2026.0174
Gemini-2.5-proSize=-2026.0174
Qwen3-VL-ThinkingSize=4B2026.0173
Kimi-VL-ThinkingSize=16B2026.0173
GPT-4oSize=-2026.0173
Qwen3-VL-InstructSize=4B2026.0172
Kimi-VL-InstructSize=16B2026.0172
Qwen2.5-VL-InstructSize=32B2026.0171
Qwen3-VL-ThinkingSize=2B2026.0169
Qwen2.5-VL-InstructSize=7B2026.0169
Molmo2Size=8B2026.0169
Molmo2Size=4B2026.0168
MiniCPM-V-4.5Size=8B2026.0168
Qwen3-VL-InstructSize=2B2026.0167
InternVL3.5Size=8B2026.0167
InternVL3.5Size=30B2026.0167
InternVL3.5Size=4B2026.0166
VideoLLaMA3Size=7B2026.0165
MiniCPM-V-4Size=4B2026.0163
Qwen2.5-VL-InstructSize=3B2026.0161
VideoChat-FlashSize=7B2026.0160
VideoLLaMA3Size=2B2026.0158
VideoChat-FlashSize=2B2026.0157
LLaVA-NeXT-VideoSize=34B2026.0153
GLM-4.6vSize=108B2026.0152
GLM-4.5vSize=108B2026.0150
LLaVA-NeXT-VideoSize=7B2026.0149
GLM-4.6v-flashSize=108B2026.0149