CreateOS AI Gateway
The unified AI gateway for every model. Send one request to one endpoint and the Gateway routes it to the cheapest available source for that model, with automatic failover. One API, one key, one bill, across models from every major provider.
Agent Hub/Support Triage
Model
Allowed by your admin · 5 models
- openai/gpt-4ovia OpenAI
- anthropic/claude-sonnetvia Anthropic
- openrouter/google/gemini-provia OpenRouter
- openrouter/mistral/mistral-largevia OpenRouter
- bedrock/meta/llama-70bvia AWS Bedrock
Instructions, knowledge and tools unchanged
Call any model from one endpoint
The Gateway is OpenAI-compatible. Point any OpenAI client or curl at the base URL, pass the model by name, and switch providers without touching your code.
export CREATEOS_LLM_API_KEY=...
curl https://createosllm.nodeops.xyz/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $CREATEOS_LLM_API_KEY" \
-d '{
"model": "claude-sonnet-4-6",
"messages": [{ "role": "user", "content": "Hello from CreateOS" }]
}'
# swap "model" to gpt-4o-mini, claude-haiku-4-5, or any supported model. Same endpoint, same key.How the Gateway works
- 01Send one requestCall one endpoint with one API key. Switch models by changing the model name, with no new SDK and no code changes.
- 02Route to the cheapest sourceThe same model is sold by more than one upstream source, often at different prices. The Gateway compares the live rate across all of them and sends each request to the cheapest. You pay that upstream token price plus a flat 2.5% platform fee, and nothing else.
- 03Fail over automaticallyUpstream health is monitored in real time. If a source degrades, requests reroute to the next cheapest available one. Built-in redundancy at no extra cost.
What the Gateway gives you
- Cheapest-price routingEvery request is priced across all available upstream sources and sent to the lowest, so you never overpay for a model.
- One API, every modelModels from every major provider through a single endpoint. Switch models by changing the model name, with no code changes.
- Automatic failoverReal-time upstream health monitoring reroutes to the next cheapest available source on an outage. Redundancy at no extra cost.
- Zero Data RetentionZDR-compliant models are available on request: the Gateway does not store your data and does not train on it.
Route across every major provider
Models from every major provider through one endpoint, priced at the cheapest available source plus a flat 2.5% platform fee. No minimums, no monthly commitment. See live model pricing and a cost calculator on the pricing page.
- Anthropic (Claude)
- OpenAI (GPT, o-series)
- Google (Gemini)
- Alibaba (Qwen)
- Zhipu AI (GLM)
- Moonshot AI (Kimi)
- MiniMax
- Mistral
- DeepSeek
Common questions
What is the CreateOS AI Gateway?
The CreateOS AI Gateway is a unified API gateway that routes every model request to the cheapest available upstream source. You reach models from every major provider through one endpoint, with one API key and one bill, and switch models by changing the model name.
How does pricing work?
You pay the base upstream token price plus a flat 2.5% platform fee. The Gateway finds the cheapest available source for every request, so you always pay the lowest rate. No hidden fees, no minimums, and no monthly commitment. Live model pricing and a cost calculator are on the pricing page.
Can I switch models without changing code?
Yes. Every model is available through the same endpoint, so switching is as simple as changing the model name in your request. No new SDK and no code changes.
What happens if a provider goes down?
The Gateway monitors upstream health in real time. If a source has an outage, requests reroute automatically to the next cheapest available source, so you get built-in redundancy at no extra cost.
One API for every model, at the cheapest price
See live model pricing, or book a demo to route your workloads through the Gateway.