A Gemini 2.5 Flash Level MLLM for Vision, Speech, and Full-Duplex Multimodal Live Streaming on Your Phone

Open-source unified multimodal model

official code for "3D Question Answering via only 2D Vision-Language Models"

Supercharge Your LLM with the Fastest KV Cache Layer

[ICCV 2025 Oral] SceneSplat - Gaussian Splatting-based Scene Understanding with Vision-Language Pretraining

Code for 3D-LLM: Injecting the 3D World into Large Language Models