cua_s1: accept image embeddings and 3D positions in native language model - #56

cua_s1: accept image embeddings and 3D positions in native language model#56

hsliuustc0106 merged 4 commits into

ThinkFlowLab:mainThinkFlowLab/system1-omni:mainfrom

Levius-Fubuki:codex/cua-native-multimodal-inputLevius-Fubuki/system1-omni:codex/cua-native-multimodal-inputCopy head branch name to clipboard