omp packages

@sentiogate/pi-provider0.1.1

SentioGate provider for the Pi coding agent: route Pi through an OpenAI-compatible LLM router with zero markup on your own keys.

hushamdev Published 3 days ago Apache-2.0 ~254 downloads/mo

Untested on omp

This is a pi package. No one has reported whether it works on omp yet. If you try it, add your result below.

Did it work for you?

Result

One report per person. A new report replaces your old one.

Reports 0

No reports yet.

Readme

@sentiogate/pi-provider

Use SentioGate as a model provider in Pi. SentioGate is an OpenAI-compatible LLM router with zero markup on your own keys. It runs on a global edge, picks a model for each request with auto, and automatically falls back to another model if one fails.

Install

pi install npm:@sentiogate/pi-provider

Or try it for one run without saving it: pi -e npm:@sentiogate/pi-provider.

Configure

  1. Create an API key at https://sentiogate.app and add your own provider keys there (or use the free community pool).
  2. Export it:
export SENTIOGATE_API_KEY=your-key
  1. Pick a model:
pi --model sentiogate/auto
pi -p "say hi" --model sentiogate/auto

Inside Pi, use /model and choose a sentiogate/... entry.

Models

Model Use
sentiogate/auto Recommended. SentioGate picks the model per request.
sentiogate/openrouter/deepseek/deepseek-chat-v3-0324 Pinned low-cost model, good for background work.
sentiogate/openrouter/google/gemini-2.5-flash Pinned fast model with image input.

Pinned models only work if your key's allowed-models list includes them. Any other model id your key allows can be added via ~/.pi/agent/models.json.

Options

Env var Default Purpose
SENTIOGATE_API_KEY (required) Your SentioGate key
SENTIOGATE_BASE_URL https://api.sentiogate.dev/v1 Override the API endpoint

License

Apache-2.0