Is Ollama 0.34 a stable release yet?

Ollama 0.34 is still a release candidate. GitHub labels both v0.34.0-rc0 from September 4, 2026 and v0.34.0-rc1 from September 5, 2026 as pre-releases, so the newest stable build remains v0.33.3, published September 3, 2026. Anyone who only wants stable versions should stay on 0.33.3 until a final 0.34.0 tag appears.

Does the ChatGPT Desktop connection work on Windows or Linux?

The Ollama 0.34 release notes say setup for ChatGPT Desktop is available from the Ollama app on macOS. No Windows or Linux path is described in the v0.34.0-rc1 notes, so on those platforms this specific connection is not listed as available. Ollama itself still installs and runs models on macOS, Windows, and Linux.

How is this different from Ollama's Claude Desktop support?

Ollama shipped Claude Desktop support on August 25, 2026, which let Anthropic's desktop app route requests through Ollama as a gateway. The ChatGPT Desktop connection in Ollama 0.34 does the same job for OpenAI's client. Together the two turn Ollama into a shared local backend for both companies' desktop apps.

What else changed in Ollama 0.34 besides the ChatGPT Desktop link?

Ollama 0.34 improves structured output performance on Apple Silicon, adds support for OpenAI-compatible client tool search, and adds response compaction. The release also fixes image handling so that images still work correctly when a response has been compacted.

Ollama · 2026-09-05 · major

Ollama 0.34 — local models now run inside ChatGPT Desktop

Ollama 0.34, published on September 5, 2026 as a release candidate, lets ChatGPT Desktop answer using models that run on your own machine. Setup happens in the Ollama macOS app. The release also speeds up structured output on Apple Silicon.

GitHub repository card for Ollama, the local open-model runtime

Ollama 0.34 wires local open models into ChatGPT Desktop, set up from the Ollama app on macOS.

Quick facts

Version	v0.34.0-rc1 (pre-release)
Released	September 5, 2026
What's new	Ollama models usable in ChatGPT Desktop
Setup	Ollama app on macOS
Last stable	v0.33.3 (September 3, 2026)
License	MIT
Language	Go

What is it?

ChatGPT Desktop can now answer using models that never leave your computer. Ollama 0.34, published on September 5, 2026 as release candidate v0.34.0-rc1, adds that connection, and you turn it on from the Ollama app on macOS. Ollama is the local model runtime written in Go, MIT licensed, that downloads and serves open models on your own hardware; it sits at 180,301 stars on GitHub.

How does it work?

The link runs through the local Ollama server, so OpenAI's desktop client sends its requests to a model process on the same machine instead of a hosted API. The same release improves structured output performance on Apple Silicon, adds support for OpenAI-compatible client tool search and response compaction, and fixes image handling so images still work through a compacted response.

Why does it matter?

Keeping the chat window and swapping the model underneath is the point. People who already work in ChatGPT Desktop can point private files and prompts at an open model without learning a new app. It also extends a pattern Ollama started on August 25, 2026 with Claude Desktop support: the project is becoming the shared local backend that other vendors' desktop apps plug into.

Who is it for?

macOS developers who want to run open models locally

Frequently asked questions

Is Ollama 0.34 a stable release yet?: Ollama 0.34 is still a release candidate. GitHub labels both v0.34.0-rc0 from September 4, 2026 and v0.34.0-rc1 from September 5, 2026 as pre-releases, so the newest stable build remains v0.33.3, published September 3, 2026. Anyone who only wants stable versions should stay on 0.33.3 until a final 0.34.0 tag appears.
Does the ChatGPT Desktop connection work on Windows or Linux?: The Ollama 0.34 release notes say setup for ChatGPT Desktop is available from the Ollama app on macOS. No Windows or Linux path is described in the v0.34.0-rc1 notes, so on those platforms this specific connection is not listed as available. Ollama itself still installs and runs models on macOS, Windows, and Linux.
How is this different from Ollama's Claude Desktop support?: Ollama shipped Claude Desktop support on August 25, 2026, which let Anthropic's desktop app route requests through Ollama as a gateway. The ChatGPT Desktop connection in Ollama 0.34 does the same job for OpenAI's client. Together the two turn Ollama into a shared local backend for both companies' desktop apps.
What else changed in Ollama 0.34 besides the ChatGPT Desktop link?: Ollama 0.34 improves structured output performance on Apple Silicon, adds support for OpenAI-compatible client tool search, and adds response compaction. The release also fixes image handling so that images still work correctly when a response has been compacted.

Try it

curl -fsSL https://ollama.com/install.sh | sh