Tracked topic
Gemma is a family of open-weight language models released by Google for text generation and related NLP tasks.
… At launch, AI Edge Gallery on macOS supports five Gemma models, including the new Gemma 4 12B and Gemma 3n for lighter workloads. These models come from Google's open‑source Gemma line, which the company designs for on‑device and self‑hosted use. …
… In another test using Gemma 4 E4B, it managed 108 tokens per second, compared to the RTX 3060's mere 76 tokens per second. It needed nearly 200W of power to do that, but when power limited to just 100W, it achieved similar results: 95 tokens on the same test using Gemma 4. …
… The app uses Google's Gemma-based speech models to convert speech to text and clean up the resulting transcript. …