@prjct.app/pi-fast-mode
Toggle the priority service tier for supported Codex and Grok models in Pi.
Package details
Install @prjct.app/pi-fast-mode from npm and Pi will load the resources declared by the package manifest.
$ pi install npm:@prjct.app/pi-fast-mode- Package
@prjct.app/pi-fast-mode- Version
0.1.3- Published
- Oct 13, 2026
- Downloads
- 344/mo · 13/wk
- Author
- jlopezlira
- License
- MIT
- Types
- extension
- Size
- 1.6 MB
- Dependencies
- 1 dependency · 1 peer
Pi manifest JSON
{
"image": "https://raw.githubusercontent.com/prjct-app/pi-fast-mode/main/docs/cover-v2.png",
"extensions": [
"./index.ts"
]
}Security note
Pi packages can execute code and influence agent behavior. Review the source before installing third-party packages.
README
pi-fast-mode
Opt into the priority service tier for supported Codex and Grok models in Pi.
Install
Requires Pi with the before_provider_request event and Node.js 22.19+:
pi install npm:@prjct.app/pi-fast-mode
Reload Pi after installation.
Use
/fast on
/fast off
/fast toggle
/fast status
/fast alone toggles the setting. It starts off and is saved per session. When enabled, supported requests receive service_tier: "priority"; unsupported models are unchanged.
The allowlist covers openai-codex models gpt-5.6-luna, gpt-5.6-sol, gpt-5.6-terra, and gpt-6-astra, plus xai models grok-4.5 and grok-4.6. Provider availability and priority-tier access depend on your account. Priority can increase quota usage or cost; it does not guarantee a particular latency.
Development
From the repository root: node scripts/build-pi.mjs pi-fast-mode.
License: MIT.
