Ollama released version 0.20.6 today, focusing on enhancements for local AI deployments. The update notably improves Gemma 4 tool calling capabilities, incorporating Google's latest post-launch fixes for enhanced reliability and performance. Additionally, parallel tool calling has seen improvements for handling streaming responses more efficiently. A new integration guide is now available for the Hermes agent, facilitating its use with Ollama. The release also includes a fix for image attachment errors within the Ollama application, further refining the user experience. These updates collectively bolster Ollama's capacity for complex, multi-turn AI interactions and integrations.