@calesennett/pi-codex-fast
pi extension that adds service_tier=priority to supported OpenAI Codex requests when fast mode is enabled.
Package details
Install @calesennett/pi-codex-fast from npm and Pi will load the resources declared by the package manifest.
$ pi install npm:@calesennett/pi-codex-fast- Package
@calesennett/pi-codex-fast- Version
0.1.5- Published
- Jul 11, 2026
- Downloads
- 447/mo · 233/wk
- Author
- calesennett
- License
- unknown
- Types
- extension
- Size
- 244.8 KB
- Dependencies
- 0 dependencies · 1 peer
Security note
Pi packages can execute code and influence agent behavior. Review the source before installing third-party packages.
README
pi-codex-fast
Fast-mode extension for pi that injects service_tier: "priority" into supported OpenAI Codex requests.
Usage
Inside pi:
/codex-fastto toggle
From CLI:
pi --fast
Persistence
The enabled/disabled state is read from pi's settings files:
- global:
$PI_CODING_AGENT_DIR/settings.json(or~/.pi/agent/settings.json) - project override:
<cwd>/.pi/settings.json
under the key pi-codex-fast.enabled.
Writes go to the global settings file.
Behavior
The extension only patches provider payloads when all of these are true:
- fast mode is enabled
- the active model is one of:
openai-codex/gpt-5.4openai-codex/gpt-5.5openai-codex/gpt-5.6-solopenai-codex/gpt-5.6-terraopenai-codex/gpt-5.6-luna
All other requests are left unchanged.
Example benchmark
A local live benchmark is available in this repository under evals/. Three paired trials per model produced:
| Model | TTFB speedup | Turn speedup | Wall speedup |
|---|---|---|---|
gpt-5.6-sol |
1.52x | 1.58x | 1.57x |
gpt-5.6-terra |
1.05x | 1.34x | 1.33x |
gpt-5.6-luna |
1.30x | 2.31x | 2.25x |