@diegopetrucci/pi-openai-fast
A pi extension that enables OpenAI Codex Fast mode for ChatGPT-auth GPT-5.4 and GPT-5.5 by injecting the priority service tier.
Package details
Install @diegopetrucci/pi-openai-fast from npm and Pi will load the resources declared by the package manifest.
$ pi install npm:@diegopetrucci/pi-openai-fast- Package
@diegopetrucci/pi-openai-fast- Version
0.1.1- Published
- May 11, 2026
- Downloads
- 381/mo · 104/wk
- Author
- diegopetrucci
- License
- MIT
- Types
- extension
- Size
- 11.3 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
openai-fast
A pi extension that enables OpenAI Codex Fast mode for ChatGPT-auth GPT-5.4 and GPT-5.5.
When active, the extension injects this into eligible OpenAI Codex request payloads:
{
"service_tier": "priority"
}
The user-facing feature is OpenAI Codex Fast mode. The wire value is priority because current Codex clients map Fast mode to the OpenAI priority service tier.
Eligibility
Fast mode is only injected when all of these are true:
- The current provider is
openai-codex. - The current API is
openai-codex-responses. - The current model is
gpt-5.4orgpt-5.5. - The provider is using ChatGPT OAuth/subscription auth, not API-key auth.
- The request payload does not already include
service_tier.
Commands
/fast
/fast status
/fast on
/fast off
/fast auto
/fast toggle
Run /fast without arguments to pick an action from a menu. /fast on and /fast off are temporary session/runtime overrides. Use /fast auto to reload and follow config defaults again.
The extension defaults to off so installing the full collection does not accidentally spend Fast-mode credits.
Config
Optional global config:
~/.pi/agent/extensions/openai-fast.json
Optional project config:
.pi/openai-fast.json
Project config overrides global config.
{
"enabled": false,
"showStatus": true
}
enabled: default Fast-mode state when there is no session override.showStatus: show a compactfaststatus when Fast mode is active for the current model.
Install
Standalone npm package
pi install npm:@diegopetrucci/pi-openai-fast
Collection package
pi install npm:@diegopetrucci/pi-extensions
GitHub package
pi install git:github.com/diegopetrucci/pi-extensions
Then reload pi:
/reload
Notes
- This extension intentionally does not affect API-key OpenAI models.
- Pi may only account Fast-mode cost correctly when the backend reports
service_tier: "priority"in the streamed response. The extension does not patch usage totals to avoid double-counting. - If pi adds first-class service-tier support later, this extension skips payloads that already contain
service_tier.