Ollama

Support for the Codex App landed in v0.24.0, bringing OpenAI's desktop coding agent to local Ollama models.

ollama launch gained a desktop coding environment. v0.24.0 added ollama launch codex-app, running OpenAI's Codex desktop app locally.¹ It includes:

• A built-in browser for loading local servers and annotating pages to request changes²
• A review mode for leaving inline comments without leaving the workspace
• Parallel worktree support with built-in git functionality

Speculative decoding arrived for Gemma 4 on Mac. v0.23.1 added Gemma 4 MTP support in the MLX runner, delivering over 2x speed gains on the gemma4:31b-coding-mtp-bf16 model for coding tasks.³

/api/show caching cut integration latency. v0.23.2 cached /api/show responses, improving median latency by ~6.7x — noticeably faster load times for tools like VS Code.⁴ Claude Desktop was also removed from ollama launch in the same release; use ollama launch claude-desktop --restore to revert.⁵

ollama launch opencode added vision model support. v0.23.4 enabled image inputs for vision models when using opencode, and fixed formatting of Claude tool results when using local image paths.⁶

A pre-release architecture shift is under active testing. The v0.30.0 release candidate series migrates the engine from GGML to direct llama.cpp, adding GGUF compatibility and using MLX for Apple Silicon inference.⁷ Known pre-release gaps: llama3.2-vision and laguna-xs.2 on Windows/Linux are not yet supported; nomic-embed-text now lowercases inputs per its model card.