OpenAI Codex with Ollama

2026-01-14 · Source: Ollama Blog · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, quick

Summary

OpenAI's Codex CLI, a tool designed to read, modify, and execute code within a working directory, now supports open models via Ollama. Released on January 15, 2026, this integration allows users to leverage models like `gpt-oss:20b` and `gpt-oss:120b`, or other open-weight alternatives, directly through the Codex interface. To get started, users install the Codex CLI via npm and then launch it with the `--oss` flag. The default model is `gpt-oss:20b`, but users can specify different models, including cloud-based options like `gpt-oss:120b-cloud`, using the `-m` flag. A key requirement for optimal performance is a large context window, with at least 32K tokens recommended.

Key takeaway

For AI Engineers and Software Engineers seeking to integrate local code generation capabilities into their workflows, this Ollama-Codex integration offers a direct path. You can now use open-weight models like `gpt-oss:20b` to automate code tasks directly from your command line, reducing reliance on proprietary APIs. Ensure your Ollama setup provides at least a 32K token context window for effective operation.

Key insights

OpenAI's Codex CLI now integrates with Ollama, enabling local execution of open-weight code generation models.

Principles

• Open models enhance developer tooling.
• Context window size impacts model performance.

Method

Install Codex CLI, then run `codex --oss` to use local Ollama models. Specify models with `-m` flag; ensure 32K+ context window.

In practice

• Use `gpt-oss:20b` for local code tasks.
• Switch to `gpt-oss:120b-cloud` for larger models.

Topics

• OpenAI Codex
• Ollama
• Code Generation
• Large Language Models
• Context Window

Best for: AI Engineer, Machine Learning Engineer, Software Engineer

Related on AIssential

• Codex + Ollama = Free Unlimited Coding AI — Ollama's integration with Codex enables free, local execution of open-source LLMs for AI coding tasks.
• OpenAI gpt-oss — OpenAI's gpt-oss models offer powerful local AI with agentic features and efficient quantization under an Apache 2.0 license.
• Tool support — Ollama now enables LLMs to interact with external tools for complex tasks via a new API feature.
• OpenAI’s Codex is now in the ChatGPT mobile app — OpenAI's Codex desktop AI tool is now controllable via the ChatGPT mobile app.
• Codex + GPT-5.5 = SUPER APP! Build and Do ANYTHING! — OpenAI's GPT 5.5 and Codex offer a superior, more efficient AI coding and automation workflow.
• Build a free writing assistant with Python and Ollama — Build a local AI typing assistant using Python, Ollama, and keyboard/clipboard control for instant typo correction.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Ollama Blog.