Ollama: Launch Hub Expands, Tokenizer APIs Proposed
Launch integrations dominated activity with Pi auto-population merged and Kimi and agy support proposed, alongside a new MiMo 2.6 model and proposed tokenize endpoints. Underneath, reliability fixes for CUDA loading, truncated tool calls, and model pulls reduce setup friction and failure modes.
Duration: PT2M21S
Episode overview
This episode is a short developer briefing from Ollama.
It explains recent repository work in plain language.
- Show: Ollama
- Published: 2026-11-04T14:01:19Z
- Audio duration: PT2M21S
Transcript excerpt
This excerpt keeps the crawler page concise. Listen to the episode or use the RSS feed for the full update.
Good morning, it's November 4th, 2026.
The throughline today is less friction: the launcher is getting smarter about models and outside tools, while inference, downloads, and builds get reliability fixes.
First, ollama launch is becoming a hub. Merged change PR 19460 means Pi now saves a starting model and auto-populates its chat-model catalog, so users no longer curate a subset by hand and can switch inside Pi. In progress, PR 19506 adds support for Kimi Code C L I through its config file, mapping local models with…
Second, new model capability. Merged PR 19400 adds MiMo version 2.6 support, with candidate M L X and G G U F models available for testing. Proposed PR 19498 would add tokenize and detokenize endpoints to convert text and token IDs without loading a model, with accurate counts for prompt evaluation. That would help…
Third, reliability hardening. Merged PR 19433 preloads bundled CUDA libraries by full path on Linux, fixing M L X load failures when host libraries mismatched. Open PRs continue that: PR 19494 would drop only truncated tool calls cut off by the token limit instead of failing the whole request with a 500, and PR…
What's next: watch whether the tokenizer…
Nearby episodes from Ollama
- Ripple on MLX and Validation Hardening
- MLX Memory Cuts and Tool-Call Fixes
- Launcher Return, API Fixes and Inference Reliability
- MLX Upgrades and OpenAI Compatibility Fixes
- Reliability Fixes and Rockchip NPU Support
- Inference Correctness and Installer Reliability
- Weekly Recap - Sampler Fidelity, Faster Drafting, and Launcher Fixes
- Sampler Defaults Fix and Memory Work