Omni-Modal LLMs Know More Than They Say
Open paper
Benchmark
Model
Rank
Results
audio-visual-question-answering-on-omnibench
OMNI-CATALYST (Qwen2.5-Omni-7B)
–
Accuracy: 51.7
audio-visual-question-answering-on-worldsense
OMNI-CATALYST (Qwen2.5-Omni-7B)
–
Accuracy: 45.7
Rank counts only results with a code link.