OpenAI published release notes for Codex version 0.166.0, introducing keyboard-driven transcript selection and adding GPT-6.1 Terra to its model catalogs.confirmed
What was released or claimed
The release notes state that version 0.166.0 adds keyboard-driven transcript selection with word, line, and page motions, integrated search, and configurable shortcuts.reported It also renders Markdown consistently in async questions, choices, answers, and transcript replies. GPT-6.1 Terra was added to the bundled, Vertex AI, and Amazon Bedrock model catalogs.reported The update includes a new `codex queue --resume` command to resume unloaded threads after queueing a message. Clearer safety-block explanations and selectable safer follow-up prompts were added to the TUI. A new `codex doctor --network-only` command provides focused configuration, connectivity, and WebSocket diagnostics.
Who it affects
Users of the OpenAI Codex terminal user interface who interact with transcripts, threads, and model catalogs are affected by these changes. Developers using macOS with non-ASCII paths or in-use installations also benefit from cleanup fixes.
What is still unknown
The release notes do not specify performance benchmarks for GPT-6.1 Terra compared to previous models.reported It is unclear how the new keyboard shortcuts interact with existing custom configurations. The impact of the V8 source-build switch on build times is not detailed.
So what
The addition of GPT-6.1 Terra expands model availability for users on Vertex AI and Amazon Bedrock.reported Improved transcript navigation may enhance usability for researchers reviewing long conversations. Preserving reasoning-token accounting helps maintain context integrity during thread operations.
What the paper shows
The release notes indicate that version 0.166.0 tests improvements in user interface navigation and thread state management.reported Reported results include consistent Markdown rendering and preserved approval settings when resuming or forking threads.reported The update fixes race conditions in app-server thread startup and shutdown. Limitations are not explicitly stated, but the notes focus on bug fixes and feature additions rather than empirical performance data. The inclusion of GPT-6.1 Terra suggests an expansion of supported models, but no accuracy or speed metrics are provided.reported
Why it matters
- Researchers using Codex can now navigate transcripts more efficiently with keyboard-driven motions.
- The addition of GPT-6.1 Terra provides access to a new model variant on major cloud platforms.
- Fixes for reasoning-token accounting may improve the reliability of long-running threads.
- Improved TUI reconnection recovery enhances stability for users with intermittent network connections.
Topics: OpenAI, Codex, GPT-6.1 Terra, terminal user interface, thread management
Updated
Ask the article
Local keyword search of this article; no generated answers.Which of these is confirmed?
Check the answer
OpenAI published release notes for Codex version 0.166.0, introducing keyboard-driven transcript selection and adding GPT-6.1 Terra to its model catalogs.
The matched primary source and announcement wording support the confirmed label.
Did this article give you the evidence you needed?
Your choice stays on this device. No public tally.Questions readers ask
What new model is supported in Codex 0.166.0?
GPT-6.1 Terra is now supported in the bundled, Vertex AI, and Amazon Bedrock model catalogs.
How does version 0.166.0 handle thread resumption?
It adds `codex queue --resume` to resume unloaded threads and preserves reasoning-token accounting across resume and fork operations.
Are there any new diagnostic tools in this release?
Yes, `codex doctor --network-only` is added for focused configuration, connectivity, and WebSocket diagnostics.
Connected stories
Reading receipt
- Version 0.166.0 adds keyboard-driven transcript selection with word, line, and page motions.
- GPT-6.1 Terra is now included in bundled, Vertex AI, and Amazon Bedrock model catalogs.
- The update preserves reasoning-token accounting across resume and fork operations to avoid premature compaction.
