Releases·Tool Release
Ollama v0.34.0 runs local models in ChatGPT Desktop, improves Apple Silicon structured output
Ollama shipped v0.34.0, letting users run Ollama models directly in ChatGPT Desktop, with setup available from the Ollama app on MacOS. The release also improves structured output performance on Apple Silicon and adds OpenAI-compatible client tool search and response compaction support.
17 engagement1 sourceSat, Sep 5, 2026, 11:49 PM
AI Overview
- Ollama v0.34.0 lets local Ollama models run directly in ChatGPT Desktop, with setup from the MacOS Ollama app
- The release improves structured output performance on Apple Silicon
- v0.34.0 adds OpenAI-compatible client tool search and response compaction, fixing images through compacted responses
Related
Releases·Tool Release
Ollama ships v0.33.x releases with MLX runner updates and Qwen3.8 Flash Next support
Ollama released three versions in a week: v0.33.0 (Aug 21), v0.33.1 (Aug 26), and v0.33.2 (Aug 27). The updates add a Claude desktop app, MLX runner improvements including structured output support and Qwen3.8 Flash Next support, plus app and proxy fixes. Practitioners running local models on Mac benefit from Metal GPU timeout fixes and faster structured output.
Ollama released three versions in a week: v0.33.0 (Aug 21), v0.33.1 (Aug 26), and v0.33.2 (Aug 27). The updates add a Claude desktop app, MLX runner improvements including structured output support and Qwen3.8 Flash Next support, plus app and proxy fixes. Practitioners running local models on Mac benefit from Metal GPU timeout fixes and faster structured output.
Updated Wed, Sep 2, 2026, 12:11 AM — v0.33.2 released Aug 27 with dark mode restore, proxy catalog-change handling, and macOS handoff sync.
04036 engagement4 sourcesgithubFri, Aug 21, 2026, 10:52 PM
Releases·Tool Release
Ollama v0.32.6 adds MLX speculative decoding, OpenAI-compatible streaming
Ollama released v0.32.6, improving Apple GPU performance via automatic speculative decoding with Qwen3.5's MTP head and aligning its streaming API with OpenAI's wire format. The update also fixes finish_reason reporting and adds cloud model fallback for kimi-k3.
Ollama released v0.32.6, improving Apple GPU performance via automatic speculative decoding with Qwen3.5's MTP head and aligning its streaming API with OpenAI's wire format. The update also fixes finish_reason reporting and adds cloud model fallback for kimi-k3.
11 engagement1 sourcegithubTue, Aug 4, 2026, 06:49 PM
Releases·Tool Release
ollama releases v0.34.1 with MLX memory and cache fixes
Ollama published v0.34.1, a patch release updating its MLX runner and llama.cpp backends. Changes focus on memory handling: the mlxrunner now evicts prefix cache snapshots and waits for evicted runners and free system memory before loading the next model. The release also raises the token repeat limit to 100 and fixes a ChatGPT model selector spacing bug in the app.
Ollama published v0.34.1, a patch release updating its MLX runner and llama.cpp backends. Changes focus on memory handling: the mlxrunner now evicts prefix cache snapshots and waits for evicted runners and free system memory before loading the next model. The release also raises the token repeat limit to 100 and fixes a ChatGPT model selector spacing bug in the app.
14 engagement1 sourcegithubMon, Sep 14, 2026, 10:14 PM
Releases·Tool Release
Ollama releases v0.32.0-rc0 with Qwen3.5 support and agent UI
Ollama released v0.32.0-rc0, adding a Qwen3.5 parser and renderer, a warning for old agent models, and a new agent UI. This update enables users to work with Qwen3.5 models and improves the agent experience.
Ollama released v0.32.0-rc0, adding a Qwen3.5 parser and renderer, a warning for old agent models, and a new agent UI. This update enables users to work with Qwen3.5 models and improves the agent experience.
10 engagement2 sourcesgithubSat, Jul 11, 2026, 01:03 AM
Products·launched
Ollama desktop app with persistent memory, encrypted vault, and model switching
A macOS app that wraps Ollama with a GUI, providing semi-persistent memory that repairs itself, an AES256 encrypted vault, and live model switching. It auto-detects installed models and allows context to carry over between different models, solving the problem of losing conversation context when switching models in Ollama.
A macOS app that wraps Ollama with a GUI, providing semi-persistent memory that repairs itself, an AES256 encrypted vault, and live model switching. It auto-detects installed models and allows context to carry over between different models, solving the problem of losing conversation context when switching models in Ollama.
73 engagement1 sourceredditSun, Jul 12, 2026, 10:25 PM