Gemma Hits 1B Downloads; VLA Demo Runs Locally on Jetson
clmt · x · 2026-08-22
Google DeepMind celebrated Gemma surpassing 1 billion downloads. Developer Asier Arranz demonstrated a Gemma 4 Vision-Language-Action (VLA) setup running locally on a NVIDIA Jetson Orin Nano Super. The system autonomously decides whether to use the webcam based on context, eliminating the need for keyword triggers. The project provides a complete Python script integrating Parakeet STT and Kokoro TTS, available for users to clone and run. The author also outlined a wishlist for future Gemma iterations, advocating for native VLA and world model capabilities.
Related event: Google's Gemma Open Models Surpass 1 Billion Downloads(13 posts)→
More from coding & agent
- Architects vs gardeners: developer argues AI's plan-maxxing obsession hurts good work — nptacek · 2026-10-16
- OpenRouter launches Armor beta to block out-of-policy agent tool calls before execution — dfinke · 2026-10-16
- Meta's coding-optimized open-weight Muse Spark Open lands on Modal with 1M context — AAAzzam · 2026-10-16
- Google rewrites its SRE book for the 10th anniversary, adding agentic-era guidance — RealGeneKim · 2026-10-16
- GRPO post-training collapses model distributions, clashing with long-horizon agents — zsakib_ · 2026-10-16
- Dev uses Claude to draw animated mascot GIFs frame-by-frame in code until usage runs out — jeremy_goldberg · 2026-10-16