Ollama: Hermes Agent Integration and Gemma4 Improvements
Ollama added Hermes agent integration through a new launch command and implemented several Gemma4 performance and quality improvements. The development team also fixed model recommendation ordering in the launch interface.
Duration: PT1M55S
Episode overview
This episode is a short developer briefing from Ollama.
It explains recent repository work in plain language.
- Show: Ollama
- Published: 2026-04-15T00:00:00Z
- Audio duration: PT1M55S
Transcript excerpt
This excerpt keeps the crawler page concise. Listen to the episode or use the RSS feed for the full update.
Good morning, this is your Ollama development briefing for April 15th, 2026.
Parth Sareen merged the major Hermes agent integration, adding support for "ollama launch hermes" with built-in installation and onboarding. This 3,000-line addition introduces a new ManagedSingleModel flow that allows Hermes to select and persist Ollama-backed models through the launcher interface. The integration…
Eva H merged a fix ensuring cloud model recommendations always appear first in the launch model picker, regardless of local model inventory. Previously, cloud recommendations were incorrectly moved to the bottom when users had local models but no cloud models installed.
Daniel Hiltgen merged an optimization to keep Gemma4 router projections in source precision, trading minimal size increase for improved model quality.
Additional development focused heavily on Gemma4 improvements. Devon Rifkin implemented dynamic rendering based on model size, with smaller models like 2B and 4B now rendering differently from larger 26B and 31B variants. The system maintains backward compatibility while enabling explicit renderer specification for…
Daniel Hiltgen contributed two performance fixes: resolving a…
Nearby episodes from Ollama
- Weekly Recap - MLX Performance & Launch Integration Expansion
- MLX Sampler Improvements
- Windows WSL Integration Simplified
- Gemma4 Enhancements and Copilot CLI Integration
- Gemma 4 MLX Support and Mixed-Precision Improvements
- Weekly Recap - Model Integration and Tooling Enhancements
- ROCm 7.2.1 Performance Update
- Gemma4 Parser Improvements