List your company

Video Reasoning on VidHalluc (test)

81.15Binary QA Accuracy (ACH)GPT-4o
6.63425.979545.32564.6705Feb 19, 2026

Evaluation results

MethodLinks
GPT-4oLLM Params=~200B2026.0281.1590.958271.58
Gemini-1.5-ProLLM Params=~200B2026.0275.2779.2583.8363.96
GraphThinkerLLM Params=7B, EVSG=true, RL=true, finetuned_on_Rextime=true2026.0266.0484.5776.3357.81
VILA1.5LLM Params=13B2026.0258.4681.8863.3335.03
QwenVL-2.5+EVSGLLM Params=7B, EVSG=true2026.0250.9283.2375.3354.58
QwenVL-2.5LLM Params=7B2026.0250.4681.9967.546.77
VideoLLaMA2LLM Params=7B2026.0250.0483.8426.1765.12
PLLaVALLM Params=13B2026.0235.376.9616.532.44
ShareGPT4VideoLLM Params=8B2026.0229.9644.7849.517.83
Video-LLAVALLM Params=7B2026.0226.8464.4527.1729.6
LLaVA-NeXT-VideoLLM Params=34B2026.0226.677.5721.3344.4
Chat-UniViLLM Params=13B2026.0223.7754.7935.529.87
Video-ChatGPTLLM Params=7B2026.029.524.5830.177.7