[FEATURE] Add an option for progressive response rendering in Claude Code

Preflight Checklist

• [x] I have searched existing requests and this feature hasn't been requested yet
• [x] This is a single feature request (not multiple features)

Problem Statement

I frequently use Claude Code in an interactive terminal. When a response is long, the current rendering can look paused even though generation is still in progress. Incremental rendering would make the application feel more responsive and would make it easier to follow the response as it is generated.

Proposed Solution

Claude Code currently appears to render responses in relatively large chunks. It would be useful to have an optional progressive rendering mode that displays generated text incrementally as it arrives, similar to the terminal experience in GitHub Copilot CLI.
Suggested behavior:
• Add a configuration option such as  streaming_display: true  or  --stream .
• Render text deltas immediately instead of waiting for larger response blocks.
• Keep the current behavior as the default for users who prefer less terminal animation.
• Do not artificially delay or generate text character by character; simply display partial text as soon as it is received.
• Keep tool calls, approvals, errors, and Markdown rendering readable.
• Ideally support disabling the animation for redirected output or non-interactive terminals.
This would make long responses feel more responsive and would provide clearer visual feedback that Claude Code is actively working.

Alternative Solutions

I currently work around this by waiting for Claude Code to finish generating a response, but this makes the terminal feel unresponsive during long answers.

I have also compared the experience with GitHub Copilot CLI, which displays response content progressively as it arrives. That makes it easier to tell that generation is still active and to follow the response in real time.

I do not want to change the model's generation behavior or add artificial delays. The desired workaround is simply to display the partial text already received from the API.

Priority

High - Significant impact on productivity

Feature Category

CLI commands and flags

Use Case Example

Example scenario:

1. I am working on a web application and ask Claude Code to inspect several files and explain a proposed change.
2. Claude Code begins generating a long response containing the analysis and implementation steps.
3. During generation, the terminal may appear paused because the response is rendered in relatively large chunks.
4. With progressive rendering enabled, I could read the response as it arrives and immediately see that Claude Code is still active.
5. I could also identify earlier whether the response is going in the wrong direction and stop it if necessary.

This would make long-running interactions feel more responsive and improve feedback during interactive terminal use

Additional Context

This request is about display behavior, not about forcing the model to generate one character at a time. The preferred behavior is to render text deltas as soon as they are received from the streaming response.

The current rendering behavior should remain the default, with progressive rendering provided as an opt-in setting or command-line option. It would also be useful for the feature to disable itself automatically when output is redirected to a file or when Claude Code is running in a non-interactive terminal.

GitHub Copilot CLI provides a similar progressive terminal experience.