Ollama v0.32.0 reports a new agent workflow and integration updates
Ollama’s v0.32.0 release notes report an interactive agent launched by running `ollama`, a renamed ChatGPT integration, a simplified integration menu, and pre-launch warnings for specified older agent-model families.
Ollama v0.32.0 reports a new interactive agent entry point, an integration rename to ChatGPT, a narrower launch-menu presentation, and warnings for several older agent-model families. These are first-party release claims and were not independently verified in the supplied record.
01
What we know now
- 01
[1] Ollama, v0.32.0 release notes, first-party release record, published 11 July 2026: https://github.com/ollama/ollama/releases/tag/v0.32.0.
- 02
The supplied packet characterizes all listed changes as first-party claims without independent verification.
02
Why this matters for Armenia
Developers in Armenia using Ollama or evaluating local-AI tooling can use this release summary to review command-line workflow and integration changes reported by the project.
03
Interactive agent experience reported when running `ollama`
The release notes describe chat, coding, web search, and work delegation capabilities.Former Codex App integration name
The notes say it is now called ChatGPT and cite `ollama launch chatgpt`.Launch-menu selection
The menu reportedly foregrounds popular integrations; others remain available through `ollama launch`.Older agent-model behavior
Listed older model families reportedly show a warning before launch continues.Reported changes in Ollama v0.32.0, based on the project’s release notes.
04
A new default interactive entry point
According to Ollama’s v0.32.0 release notes, invoking `ollama` now starts an interactive agent experience. The project presents this as a developer-facing environment for working with models and handling tasks such as coding, web search, and delegated work.
The supplied material does not provide independent testing, details on the underlying agent configuration, or information about availability across platforms.
- Running `ollama` is reported to open an interactive agent experience rather than only serving as a command entry point.
- The experience is described as supporting model chat, coding, web search, and delegation of work.
05
ChatGPT integration naming changes
The release notes say that the integration previously called Codex App is now named ChatGPT. For users who relied on the earlier label, the recorded command path is `ollama launch chatgpt`, with `--restore` described as a way to return to the usual ChatGPT profile.
The supplied source does not explain migration behavior beyond that stated rename and restore option, so users should validate existing workflows locally.
- The Codex App integration is reported to have been renamed ChatGPT.
- The release notes identify `ollama launch chatgpt` as the launch command.
- The notes say `--restore` returns to the usual ChatGPT profile.
06
A more selective integration menu
Ollama says it has simplified integration selection by limiting the visible launch menu to its most popular integrations. This appears to change presentation rather than eliminate access, because the notes say other integrations can still be reached through `ollama launch`.
Since the supplied record does not define the popularity criterion or list the integrations affected, teams with less common integrations may want to confirm the revised navigation before updating internal guidance.
- The `ollama launch` menu reportedly shows only integrations designated as most popular.
- Other integrations are said to remain accessible through `ollama launch`.
- The supplied record does not identify which integrations receive prominent menu placement.
07
Warnings for older agent models
The release notes say Ollama now prompts with a deprecation warning before launching several older agent-model families. The named groups include CodeLlama, Qwen2.5 variants, Llama 3.x, Mistral, StarCoder, and base DeepSeek-R1 tags.
This is a warning-oriented change in the supplied account. It does not establish that the listed models can no longer be used, nor does it provide a replacement-model recommendation.
- Warnings are reported for CodeLlama, Qwen2.5 and Qwen2.5-coder, Llama 3.x, Mistral, StarCoder, and base DeepSeek-R1 tags.
- The warning is reported to appear before `ollama launch` proceeds.
- The source describes a deprecation warning, not removal of the models.
08
Practical checks before upgrading
The supplied release record is first-party material. Teams can test the reported changes in a non-production environment before changing established workflows.
- 01
Run `ollama` and assess whether the interactive agent flow fits your development workflow.
- 02
If using the former Codex App entry, test the reported `ollama launch chatgpt` path and the `--restore` option.
- 03
Check whether an integration needed by your team remains easy to reach under the streamlined launch menu.
- 04
Identify any use of the listed older model families and review the new warning behavior.
09
Limits of this edition
This article relies solely on Ollama’s first-party v0.32.0 release notes; no independent verification was supplied.
The supplied record does not specify platform availability, upgrade requirements, or compatibility details.
The record describes warnings for older models, not their removal or a loss of support.
SRC
Source desk
Direct links to the material behind this selection. Seeing the source matters as much as reading the synthesis.
Keep exploring
21 Aug 2026
Ollama v0.32.15 adds onboarding and reports faster response starts
Ollama’s official v0.32.15 release focuses on a smoother first launch, a reported reduction in time-to-first-token, and a fix for a streaming parser-error condition that could stall chat and generate. Users should treat the latency result as a vendor benchmark and validate it locally. [1]
19 Aug 2026
Ollama v0.32.12 adds Qwen 3.8 27B support
Ollama v0.32.12 adds support for Qwen 3.8 27B. The release provides a standard run command and an MLX variant that Ollama says is optimized for Apple Silicon devices, while leaving performance benchmarks and hardware requirements unspecified.
22 Aug 2026
Ollama v0.32.12 adds Qwen 3.8 27B support
Ollama’s official v0.32.12 release adds stated support for Qwen 3.8 27B through `qwen3.8:27b` and introduces `qwen3.8:27b-mlx`, an Apple Silicon-focused variant. The stated optimization benefits have not been independently corroborated in the supplied material.