HAIMAKER DEVELOPER RESOURCES
One OpenAI-compatible endpoint for 200+ models. Point your SDK at https://api.haimaker.ai/v1 and every model in the catalog works with no other code changes.
QUICKSTART
The gateway speaks the OpenAI protocol, so any OpenAI SDK works as-is: change base_url, use your Haimaker key, and name any model from the catalog. Set the model to haimaker/auto and the auto-router picks the optimal model per request by cost, speed, or capability.
from openai import OpenAI
client = OpenAI(
base_url="https://api.haimaker.ai/v1",
api_key="YOUR_HAIMAKER_API_KEY",
)
response = client.chat.completions.create(
model="haimaker/auto", # or any model from the catalog
messages=[{"role": "user", "content": "Hello"}],
)API DOCUMENTATION
Getting started
Authentication, supported protocols, and a first Chat Completions request.
Chat Completions
Request parameters and response shape for POST /v1/chat/completions.
Responses API
POST /v1/responses for catalog models that run in responses mode.
Anthropic Messages
POST /v1/messages — Anthropic-format requests against any routed model.
Gemini generateContent
Gemini-format requests through the same gateway.
Auto-router
Set model to haimaker/auto and each request routes to the optimal model.
Key management
Create and manage API keys programmatically.
Connect CLI
npx @haimaker/connect writes your coding agent’s native config, then exits.
MACHINE-READABLE RESOURCES
Every page on this site also negotiates markdown: send Accept: text/markdown to any URL, or append .md, and you get the content without the layout.
OpenAPI spec
OpenAPI 3.1 schema for every supported inference protocol.
Model catalog (JSON)
Current model names, modes, prices, providers, and capabilities.
Model catalog (browse)
Compare 200+ models by cost, speed, and capability.
llms.txt
Site index for AI agents — every page and blog post with descriptions.
llms-full.txt
Extended index with complete blog post content.
Docs llms.txt
Index of the API documentation for AI agents.