NVIDIA Jetson Thor Runs Qwen3.5 VLM Locally in Real-Time Demo
chrismatthieu · x · 2026-09-02
A demo featuring NVIDIA Jetson Thor and RealSense has successfully run the Ollama Qwen 3.5 4B VLM model locally. The system describes what it sees, including depth, every 350ms, showcasing the combination of massive AI compute with real-time 3D perception.
More from Embodied
- Isaac 0.5 plays tic tac toe, adapting to environment and opponents — chris_j_paxton · 2026-09-02
- User claims Waymo feels safer at intersections than human drivers — matt_slotnick · 2026-09-02
- Satire on Waymo criticism: Negative ads focus on "car didn't hit child" — ATTlKA · 2026-09-02
- Robot choir: each Microduck keeps its own lifelong audio identity — huggingface · 2026-09-02
- No Priors: Max Hodak breaks down PRIMA, the first device to restore vision — No Priors · 2026-09-02
- Stop building remote-controlled robots; sim2real is the key to success — _Stocko_ · 2026-09-02