LLM
9 October 2026
OpenAI's Codex introduces next message feature in beta version
N
New-ZZZ desk
X @OpenAIDevs · 2 days ago
OpenAI introduced a new feature in the beta version of Codex: "composer predictions." It analyzes your current conversation and communication style to suggest the most probable next message. This significantly speeds up the text writing process and improves interaction with the model.
The feature is available in the Codex desktop application for Pro users and is activated by pressing the Tab key. Users can easily disable this feature in the settings if they do not need it. This is another improvement aimed at increasing the efficiency of working with large language models.
Why it matters
- —Increases speed: the function automatically suggests the next text fragment, minimizing manual input.
- —Improves user experience: makes the dialogue process with AI more natural and intuitive.
- —Expands LLM capabilities: adds "composer" features to the working tool.
Key facts
- The function is called "composer predictions".
- Available in beta version for paid Pro users.
- Suggests text based on the context of the current conversation.
- To accept the suggestion, it is enough to press the Tab key.
The full text is in the original source. Here we provide a brief summary and key facts.