Skip to main content
big-AGI is a feature-rich, multi-provider AI front-end that supports adding up to five custom OpenAI-compatible endpoints. Point one at https://api.tokios.com/v1 with your sk-tok-… key, and big-AGI talks to a model running on your own hardware alongside any other providers you’ve configured.

​
Prerequisites

Connector paired

Install and pair tokios-connector next to the model you want big-AGI to reach.

Model registered

Register a deployment (public name) for that model.

API key created

Create an sk-tok-… key, optionally scoped to that deployment.

Supported backends

Confirm your local backend (Ollama, llama.cpp, vLLM, LM Studio) is running.

​
Add Tokios as a custom endpoint

1

Open the models configuration

Open big-AGI’s Models settings and choose to add a source. Select the OpenAI-compatible vendor (the option for a custom OpenAI-style endpoint). The exact menu path and labels vary by big-AGI version — confirm against the version you’re running.
2

Set the API host and key

Enter the Tokios endpoint and your key:
Include /v1 in the host.
3

Load your deployment

big-AGI auto-detects models from the endpoint. Because Tokios implements GET /v1/models, your registered deployments (for example gemma-tunnel) appear for selection. If a model isn’t listed, confirm your key is scoped to that deployment on the Keys tab.
For a self-hosted big-AGI, you can set the same values with environment variables — OPENAI_API_HOST=https://api.tokios.com/v1 and OPENAI_API_KEY=sk-tok-YOUR_KEY. Never commit a real key into version control or a shared config.

​
One front-end, models from anywhere

big-AGI is built to talk to many providers at once. Adding Tokios as one of its custom OpenAI-compatible endpoints slots your own local model in next to everything else — the connector next to that model dials out to Tokios, and big-AGI reaches it at https://api.tokios.com/v1 without an inbound port or VPN.

​
Troubleshooting

The API key big-AGI sent is missing, malformed, or has been revoked. Confirm you used the full sk-tok-… key with no extra whitespace, and check its status on the Keys tab.
big-AGI populates models from GET /v1/models. Make sure you’ve registered at least one deployment on the Models tab and that your key’s scope includes it. Confirm the host is https://api.tokios.com/v1.
The connector for that deployment is offline or at its concurrency cap. Check the Connectors tab — the connector needs to show as Online, and the machine running it needs to still be up and reachable.

​
Next steps

Compatible clients

See base URL and key setup for other OpenAI- and Anthropic-compatible clients.

API keys

Create, scope, and rotate sk-tok-… keys for teammates and agents.