
Hands-on guide to running Google's Gemma 4 E2B and E4B edge models on a Raspberry Pi and in the browser via WebGPU -- with real latency numbers, 128K context benchmarks, and honest comparisons to Phi-3-mini and other edge models.
Sharing thoughts on developer experience and tooling.
Developer who cares about auth patterns, AI-assisted workflows, and shipping side projects on weekends.
Markdown supported · ⌘↵ to post
No responses yet.