@pi-plugins/fast-mode

Fast service-tier extension for pi-agent.

Packages

Package details

extension

Install @pi-plugins/fast-mode from npm and Pi will load the resources declared by the package manifest.

$ pi install npm:@pi-plugins/fast-mode
Package
@pi-plugins/fast-mode
Version
0.1.8
Published
Jul 23, 2026
Downloads
1,601/mo · 331/wk
Author
kedom1337
License
MIT
Types
extension
Size
27.7 KB
Dependencies
2 dependencies · 1 peer
Pi manifest JSON
{
  "extensions": [
    "./dist/index.mjs"
  ]
}

Security note

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

README

@pi-plugins/fast-mode

A pi-agent extension that requests a provider's fast (priority) service tier for configured models — faster inference at a higher token cost.

Fast mode is a per-model opt-in: you decide which provider/model-id pairs it applies to, and a single session toggle turns it on and off. Requests for any other model are left untouched.

Install

pi install npm:@pi-plugins/fast-mode

For a one-off run without adding it to settings:

pi -e npm:@pi-plugins/fast-mode

For local development, load it straight from this directory:

pi -e ./plugins/fast-mode

Supported providers

Provider Mechanism
openai service_tier: "priority" payload
openai-codex service_tier: "priority" payload

Usage

/fast          # toggle fast mode for this session
/fast on       # enable
/fast off      # disable
/fast status   # show the current state and why it is (in)active

Start a session with fast mode already enabled:

pi --fast

While fast mode is on and the current model is configured for it, a dim [fast mode] indicator is shown at the right of a status line right above the editor, shared with other pi-plugins extensions (e.g. @pi-plugins/speed renders its measurements at the left of the same line).

Configuration

Optional config file at <agent-dir>/extensions/fast-mode.json (typically ~/.pi/agent/extensions/fast-mode.json):

{
  "enabled": false,
  "models": [
    "openai/gpt-5.4",
    "openai/gpt-5.5",
    "openai/gpt-5.6-sol",
    "openai/gpt-5.6-terra",
    "openai/gpt-5.6-luna",
    "openai-codex/gpt-5.4",
    "openai-codex/gpt-5.5",
    "openai-codex/gpt-5.6-sol",
    "openai-codex/gpt-5.6-terra",
    "openai-codex/gpt-5.6-luna"
  ],
  "showStatus": true
}
  • enabled: fast-mode state at session start (/fast overrides it for the session; the toggle is not written back to the file).
  • models: the provider/model-id keys fast mode applies to. The defaults above are used when the file is absent. Models whose API has no fast-mode support are ignored.
  • showStatus: show the [fast mode] indicator above the editor while active.

Notes

  • The priority service tier bills at a higher rate than the default tier — check your provider's pricing before leaving it enabled.
  • If another extension already set a service_tier on the payload, this extension leaves it alone.