Ollama: Weekly Recap - Gemma4 Integration & Audio Support
This week featured the major addition of Gemma4 model support with comprehensive text, vision, and audio capabilities, alongside critical bug fixes for ROCm builds and tokenization improvements.
Duration: PT2M33S
Episode overview
This episode is a short developer briefing from Ollama.
It explains recent repository work in plain language.
- Show: Ollama
- Published: 2026-04-06T00:00:00Z
- Audio duration: PT2M33S
Transcript excerpt
This excerpt keeps the crawler page concise. Listen to the episode or use the RSS feed for the full update.
Good morning, I'm your host with the Ollama weekly recap for March 30th through April 6th, 2026.
Zero pull requests were merged, with 15 additional commits this week.
**Major Features**
The headline this week is comprehensive Gemma4 model support. Daniel Hiltgen delivered a massive implementation adding the complete Gemma 4 model family - including E2B, E4B, 26B MoE, and 31B Dense variants. The integration includes text processing with sliding window attention and per-layer embeddings, vision…
The audio implementation is particularly robust, featuring microphone recording across platforms - AVFoundation on macOS, PulseAudio and ALSA on Linux, and WASAPI on Windows. Users can now transcribe audio through a new dedicated transcribe command and interactive voice chat modes.
Parth Sareen added a launch page for integration shortcuts, while the team defaulted the app home view back to new chat for improved user experience.