@hk-vk/pi-openai-fast

Enable Fast mode for Pi's OpenAI providers.

Packages

Package details

extension

Install @hk-vk/pi-openai-fast from npm and Pi will load the resources declared by the package manifest.

npm repo report
$ pi install npm:@hk-vk/pi-openai-fast
Package
@hk-vk/pi-openai-fast
Version
0.1.1
Published
Oct 21, 2026
Downloads
158/mo · 17/wk
Author
hk-vk
License
MIT
Types
extension
Size
10.2 KB
Dependencies
0 dependencies · 1 peer
Pi manifest JSON
{
  "extensions": [
    "./index.ts"
  ]
}

Security note

Pi packages can execute code and influence agent behavior. Review the source before installing third-party packages.

README

@hk-vk/pi-openai-fast

Enable Fast mode for models using Pi's openai and openai-codex providers.

Forked from Diego Petrucci's pi-extensions.

Features

  • Toggle Fast mode for the current session with /fast.
  • Show fast in the footer when the selected model is eligible and Fast mode is on.
  • Applies to eligible models by provider and API, without a hardcoded model-ID list.
  • Set the default and status indicator globally or per trusted project.

Supported APIs

  • openai: Responses and Completions
  • openai-codex: Responses with ChatGPT OAuth

Install

pi install npm:@hk-vk/pi-openai-fast

Run /reload, then /fast to enable it for the session. Fast mode is off by default.

Configuration

Global config: ~/.pi/agent/extensions/openai-fast.json

Project config: <project>/.pi/openai-fast.json

{
  "enabled": false,
  "showStatus": true
}

Project config overrides global config in trusted projects.