Remote Providers

You do not need a local GPU to participate. Any OpenAI-compatible cloud API can be used as your LLM endpoint.

How It Works

When you join a room with a remote provider, the participant runtime:

1. probes the provider endpoint locally
2. registers itself with the hub
3. opens a tunnel back to the hub
4. forwards tunnel requests to the provider endpoint using local auth headers

This means the hub does not need your provider credentials as part of public participant state, and your provider endpoint does not need to be exposed to the rest of the network.

Supported protocols:

Protocol	Endpoint	Best For
Responses API	/v1/responses	preferred default when the provider supports it
Chat Completions	/v1/chat/completions	compatibility for providers and tools that still require it

Join With CLI

For automation, prefer --header-env so secrets do not end up in shell history.

export OPENROUTER_AUTH="Bearer sk-or-..."

gambi participant join --room ABC123 \
  --participant-id my-openrouter \
  --endpoint https://openrouter.ai/api \
  --model meta-llama/llama-3.1-8b-instruct:free \
  --nickname my-openrouter \
  --header-env Authorization=OPENROUTER_AUTH

You can also send extra provider-specific headers:

gambi participant join --room ABC123 \
  --participant-id my-openrouter \
  --endpoint https://openrouter.ai/api \
  --model meta-llama/llama-3.1-8b-instruct:free \
  --nickname my-openrouter \
  --header-env Authorization=OPENROUTER_AUTH \
  --header "HTTP-Referer=https://my-app.example"

Join Through The SDK

If you are building your own participant runtime, use createParticipantSession():

import { createParticipantSession } from "gambi-sdk";

await createParticipantSession({
  hubUrl: "http://localhost:3000",
  roomCode: "ABC123",
  participantId: "my-openrouter",
  nickname: "my-openrouter",
  endpoint: "https://openrouter.ai/api",
  model: "meta-llama/llama-3.1-8b-instruct:free",
  authHeaders: {
    Authorization: `Bearer ${process.env.OPENROUTER_TOKEN}`,
  },
});

authHeaders stay local to the participant runtime. They are applied only when that runtime talks to the provider endpoint.

Popular Providers

Provider	Base URL	Free Models?
OpenRouter	https://openrouter.ai/api	yes
Together AI	https://api.together.xyz	free tier
Groq	https://api.groq.com/openai	free tier
Fireworks	https://api.fireworks.ai/inference	free tier
OpenAI	https://api.openai.com	no

Cost Considerations

When you join with a cloud provider:

• your API key is used for every request routed to you
• you pay for the tokens consumed
• other participants do not see your provider credentials

Choose a model you are comfortable paying for, or use free-tier models for experimentation.

Security Notes

• Provider auth headers should remain on the participant runtime.
• The tunnel-first model avoids publishing the provider endpoint or its credentials to the rest of the network just to participate.
• If your hub is reachable outside your LAN, put it behind HTTPS or a reverse proxy and apply proper access control before using cloud provider credentials with it.

When To Use Which Protocol

• Responses API: preferred default
• Chat Completions: compatibility path

The hub can adapt between the two protocols when needed, but new integrations should prefer Responses.