gemma-video-agent. Multi-stage video tracking pipeline on TPU using Gemma 4 multimodal (served via vLLM) + SAM 3 + RF-DETR + IoU tracker. Submitted for the Google AI GDE TPU Sprint 2026.

github.com/PromtEngineer/gemma-video-agent

Vaya's read on this project

Problem, audience, market, and the verdict — sign in to see it.

Updates

No recent activity.