Charlie Marsh Accelerates Codex CLI Launch by 25 Times
Charlie Marsh, CEO of Astral, announced a significant performance improvement for Codex CLI, which now launches 25 times faster. This achievement was made possible by rewriting the code from TypeScript to Rust.
Charlie Marsh Accelerates Codex CLI Launch by 25 Times
Charlie Marsh is the founder and CEO of Astral, a company that has developed uv, Ruff, ty, and many other Python tools.
In March of this year, Astral was acquired by OpenAI, and now Charlie works in the Codex division.
In the latest release of Codex, he revamped the command line launch, making it 25 times faster.
The CLI launch process (i.e., when you invoke the codex command) includes loading configurations, authentication, establishing connections, preparing the environment, etc. Previously, this chain took a noticeable amount of time, but now everything happens literally instantly (try it).
The specific reasons for such a leap are not disclosed. It is known that Codex CLI was previously rewritten from TypeScript to Rust, and Charlie is a true guru in it. He promises that "he can speed it up even more."
Why it matters
AnalysisThis is important because the significant performance improvement of Codex CLI can greatly enhance the efficiency of developers using this tool. Reducing the launch time for commands can accelerate the entire development process.
Discuss in community
Share your questions and insights with developers
Related in Ecosystem
Practical guides, core terminology, and agent skills
ChatGPT Updates: Plugin Development and Codex Workflows
Comprehensive engineering guide to creating, packaging, testing, and submitting plugins for ChatGPT and Codex: manifest specification, MCP servers, lifecycle hooks, custom marketplaces, and OpenAI review.
Part II. How to Work with Codex + Optimization/Security/Team
Codex is not just code autocomplete in a chat; it is a full-fledged engineering agent: it reads the repository, plans changes, edits files, runs tests, reviews its own diffs, and c...
CrewAI
One of the most popular Python frameworks for creating autonomous teams of agents, based on role distribution of responsibilities, tools, and task delegation.
LangChain Framework
The most popular open-source framework (available in Python and TypeScript) for building applications based on large language models. It provides hundreds of ready-made modules to integrate LLMs with documents, vector databases, API tools, and memory (Chains).