Models, usage & gateway
One developer station for platform models and bring-your-own providers — transparent sell rates, monthly metered billing, spend caps, and an OpenAI-compatible API on api.nodeda.com. Available on every plan.
Capabilities
Grand central for your LLM traffic
Catalog, billing, BYO routing, and a single API surface — without a separate AI vendor to wire up for every app.
Model catalog
Pick from platform models with published sell rates. Enable the models your org needs, set a default, and keep the rest off the menu for your API keys.
Metered monthly billing
Platform usage bills on tokens each calendar month — provider list price plus a clear 15% platform margin. No prepaid packs to top up.
Bring your own provider
Route through your own OpenAI-compatible endpoint and key. You pay your provider for tokens; NodeDa Work charges only a small gateway routing fee.
Spend caps
Optional monthly USD caps so teams can experiment without surprise invoices. Organization admins turn the hub on or off and choose routing mode from one setup panel.
OpenAI-compatible API + NodeDa Kit
POST chat completions on api.nodeda.com, or call client.llmHub from NodeDa Kit (Apple 1.2.1+, Android 1.3+, Flutter 1.2.1+) with the same llm:invoke-scoped developer key.
Scoped access
Only org admins and members with API access see LLM Hub. Keys need the llm:invoke scope — tokens never expand what their owner could do signed in.
How it works
From setup to first completion
Organization admins configure once. Developers call the gateway with a scoped key.
Enable the hub
Open Developer → LLM Hub, turn the hub on, choose platform, BYO, or hybrid routing, and set an optional spend cap.
Pick models & keys
Enable catalog models you want to expose. Mint a developer API key with llm:invoke, or connect a BYO key under Org Settings → Custom LLM.
Call & meter
Use NodeDa Kit (client.llmHub) or POST chat completions. Usage and estimated bill roll up in the hub overview for the current month.
FAQ
Frequently asked questions
- What is LLM Hub?
- LLM Hub is the NodeDa Work developer gateway for large language models. Browse a model catalog with transparent sell rates, choose platform models or bring your own provider key, set a monthly spend cap, and call an OpenAI-compatible chat completions API on api.nodeda.com. Usage is metered and billed monthly.
- How is platform model usage priced?
- Platform models are billed on metered token usage each calendar month. Customer rates are the provider list price plus a 15% platform margin. There is no prepaid token pack — you pay for what you invoke.
- What if I bring my own LLM provider?
- Bring-your-own (BYO) traffic uses the API key you configure under Org Settings → Custom LLM. You pay your provider directly for model tokens. NodeDa Work charges only a gateway routing fee for successful requests (a small per-request fee plus a per-million-token metering fee).
- Who can use LLM Hub?
- LLM Hub is available on every plan. Organization admins and members with API access can open Developer → LLM Hub, manage configuration, and mint API keys with the llm:invoke scope.
- Where do I call the API?
- Prefer NodeDa Kit (client.llmHub.createChatCompletion or client.llmHub.chat) with scope llm:invoke — Apple 1.2.1+, Android 1.3+, Flutter 1.2.1+. Or call the unified NodeDa API at https://api.nodeda.com — POST /v1/organizations/{orgId}/llm/chat/completions with Authorization: Bearer <developer API key>. Full reference: Developer → API reference → LLM Hub and NodeDa Kit.
- Does NodeDa Kit support LLM Hub?
- Yes on all three official SDKs: Apple NodeDa Kit 1.2.1+, Android Kit 1.3+, and Flutter Kit 1.2.1+ via client.llmHub. Prefer omitting model so Hub routing defaults apply; use LLMHubModelID for Nrova catalog ids. Read assistant text from ChatCompletionResponse.firstContent.
Ready to try LLM Hub?
Beta on every plan for org admins and members with API access. Start in Developer → LLM Hub after you sign up.