Codex can copy a conversation into a new session and show estimated costs

Correction, October 11, 2026: Session forking preserves conversation history, not an isolated copy of project files. The fork command requires a session ID or thread name.

What happened

OpenAI shipped Codex 0.148.0 on August 18. Codex is OpenAI’s command-line coding agent: you run it in a terminal and it reads, writes, and executes code on your machine, similar to Claude Code.

Two changes stand out. A session fork copies your conversation and its context into a new thread. You can also archive or restore past sessions from the TUI (the terminal interface Codex runs in) resume picker.

Second, Codex can now show “estimated thread credits or cost” directly in /status, in the terminal’s status line, and in the terminal window’s title bar. Credits are OpenAI’s unit for metering how much of a plan’s usage allowance a session has consumed, so this is a running estimate of what the current session is using or costing. The release notes limit it to “eligible workspaces.” Check whether the estimate appears in your workspace before relying on it.

Why it matters

Forking lets you explore another approach without losing the conversation that led to it. Say Codex has spent twenty minutes building context on your codebase. You can copy that history into a new thread and continue with a different prompt. The original conversation stays available.

The files are a separate concern. A conversation fork does not create an isolated project copy. Both sessions can still edit the same working files. Unlike Base44’s app Branches, which separate app changes from the live version, a session fork preserves conversational context. Use a separate checkout when the code experiment needs isolation.

Session cost estimates make ongoing usage easier to see while an agent works. The displayed estimate is useful for monitoring, though it does not replace billing records. Availability depends on workspace eligibility, so confirm the estimate appears before budgeting around it.

Who should care

Anyone running longer Codex sessions where a wrong turn costs real rework time, and anyone who has been surprised by how much a session cost after the fact. If you need to monitor ongoing spend, check whether your workspace shows the estimate.

What builders should do next

Run codex exec fork <SESSION_ID> [PROMPT] from your terminal. SESSION_ID can be an existing session ID or thread name; PROMPT is optional. The implementation preserves the original conversation, but switching conversations does not undo file edits.

For risky experiments, use a separate checkout, a working folder with its own project files. Inspect changes before returning to earlier work.

To check whether you have cost visibility, look at your terminal’s title bar or window tab while running Codex, or type /status inside a session.

Update with npm install -g @openai/codex@latest if you installed through npm, or run codex update to use Codex’s built-in updater. Confirm you’re on the new version with codex --version.

End of article