API reference

modelhaus speaks the OpenAI API. Point your existing client at our base URL, use your modelhaus key, and every sovereign and aggregated model is one call away.

Quickstart

The base URL for all API calls is:

Grab a key from your dashboard (create one at sign up), then make your first call:

# Your first completion
curl https://modelhaus.ai/v1/chat/completions \
  -H "Authorization: Bearer $MODELHAUS_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-5-2",
    "messages": [{"role":"user","content":"Say hello in one line."}]
  }'

Authentication

Authenticate every request with a bearer token in the Authorization header:

Authorization: Bearer $MODELHAUS_KEY

Keys are created and revoked in your dashboard. Each key draws down your prepaid balance; usage is metered per request. Keep keys server-side - never ship them in client code.

OpenAI compatibility

modelhaus implements the OpenAI Chat Completions and Models endpoints. Any OpenAI-compatible SDK, framework, or tool works by overriding the base URL:

Chat completions

Create a model response for a conversation. Core parameters:

Example response (non-streamed):

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "model": "glm-5-2",
  "choices": [{
    "index": 0,
    "message": {"role":"assistant","content":"Hello!"},
    "finish_reason": "stop"
  }],
  "usage": {"prompt_tokens":12,"completion_tokens":3,"total_tokens":15}
}

Streaming

Set "stream": true to receive Server-Sent Events. Each event is a chat.completion.chunk with a token delta; the stream ends with data: [DONE].

curl -N https://modelhaus.ai/v1/chat/completions \
  -H "Authorization: Bearer $MODELHAUS_KEY" \
  -d '{"model":"glm-5-2","stream":true,"messages":[{"role":"user","content":"Count to 3"}]}'

# data: {"choices":[{"delta":{"content":"1"}}]}
# data: {"choices":[{"delta":{"content":", 2"}}]}
# data: [DONE]

List models

Returns the models available to your key in OpenAI list format. Use it to discover ids at runtime.

curl https://modelhaus.ai/v1/models \
  -H "Authorization: Bearer $MODELHAUS_KEY"

Image generation

Generate images with our sovereign image models - flux-schnell and sdxl - hosted in Canada and billed per image. OpenAI images format.

curl https://modelhaus.ai/v1/images/generations \
  -H "Authorization: Bearer $MODELHAUS_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"flux-schnell","prompt":"a red fox in snow","n":1,"size":"1024x1024"}'

The response is OpenAI-compatible, returning base64-encoded images:

{
  "created": 1789132146,
  "model": "flux-schnell",
  "data": [{ "b64_json": "iVBORw0KGgo..." }]
}

SDKs & frameworks

Because modelhaus is OpenAI-compatible, the official SDKs and popular frameworks work unchanged - just set the base URL.

from openai import OpenAI

client = OpenAI(
    base_url="https://modelhaus.ai/v1",
    api_key="$MODELHAUS_KEY",
)
resp = client.chat.completions.create(
    model="glm-5-2",
    messages=[{"role":"user","content":"Hello!"}],
)
print(resp.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://modelhaus.ai/v1",
  apiKey: process.env.MODELHAUS_KEY,
});
const r = await client.chat.completions.create({
  model: "glm-5-2",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(r.choices[0].message.content);

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    base_url="https://modelhaus.ai/v1",
    api_key="$MODELHAUS_KEY",
    model="glm-5-2",
)
print(llm.invoke("Hello!").content)

Errors

Errors use the OpenAI error shape with standard HTTP status codes:

{ "error": { "message": "...", "type": "invalid_request_error" } }

Rate limits

Default limits scale with your balance and usage history and are generous for normal development. If you hit 429, back off and retry. For dedicated throughput, reserved capacity, or a private deployment, reach out.