Ollama: Weekly Recap - Phonon Voice Goes GA and MLX Speed Gains
Ollama shipped GA Phonon text-to-speech with voice catalog and text normalization, alongside faster MLX prefill for Gemma, Flint and Laguna. Streaming error handling, tool-call parsing, and launcher context fixes rounded out a reliability-focused week.
Duration: PT3M29S
Episode overview
This episode is a short developer briefing from Ollama.
It explains recent repository work in plain language.
- Show: Ollama
- Published: 2026-10-26T09:09:48Z
- Audio duration: PT3M29S
Transcript excerpt
This excerpt keeps the crawler page concise. Listen to the episode or use the RSS feed for the full update.
Hello, this is your Ollama briefing for October 19th through October 26th.
50 pull request activity items, 30 additional commits this week.
The lead story: Ollama shipped a complete text to speech stack while making Apple silicon models faster and streaming APIs more reliable.
First, voice. Four connected changes bring Gradium Phonon to general availability. Pull request 19166 adds text to speech through the existing MLX runner, with an OpenAI compatible speech endpoint and command line playback. Pull request 19227 replaces the prerelease checkpoint format with the GA format, including…
Second, MLX performance and model coverage. Pull request 19291 bumps the MLX libraries to use native sliding window attention, speeding up prompt processing for Gemma, Flint and Laguna models, with the largest gains on long prompts. Pull request 19196 adds support for the D3 decision model family, plus VRAM…
Third, correctness across tool calling, streaming, and launchers. For tool use, pull requests 19226 and 19274 recover Gemma 4 calls with trailing noise or unclosed lists, pull request 19228 preserves literal function tags in Qwen Coder parameters, and open pull request 19252 would stop integers…
Nearby episodes from Ollama
- OpenAI Fidelity and Runner Stability
- LFM3 Speech, Video and Reliability Fixes
- Scheduler Memory Fixes and Faster Drafting
- Mac Prefill Gains and Scheduler Fixes
- Gemma 4 Reliability and Runner Parity
- Tool Call Fixes and Phonon Voice Update
- Phonon Voice Goes Stable and API Failures Get Fixed
- Silent Failure Fixes and Mac Expansion