API Examples
Connect to Huddle01 AI inference using OpenAI-compatible APIs and SDKs.
This page shows exactly how to connect and run inference on Huddle01.
Connection details
- Base URL:
https://gru.huddle01.io/v1 - API key header:
Authorization: Bearer <HUDDLE_API_KEY> - Content type:
application/json
OpenAPI-style spec
openapi: 3.0.3
info:
title: Huddle01 AI Inference API
version: "1.0.0"
servers:
- url: https://gru.huddle01.io/v1
paths:
/chat/completions:
post:
summary: Generate model response from chat messages
security:
- bearerAuth: []
requestBody:
required: true
content:
application/json:
schema:
type: object
required: [model, messages]
properties:
model:
type: string
example: gpt-4o-mini
messages:
type: array
items:
type: object
properties:
role:
type: string
enum: [system, user, assistant]
content:
type: string
responses:
"200":
description: Successful inference response
components:
securitySchemes:
bearerAuth:
type: http
scheme: bearer
bearerFormat: API KeycURL example
curl -X POST "https://gru.huddle01.io/v1/chat/completions" \
-H "Authorization: Bearer $HUDDLE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-4o-mini",
"messages": [
{ "role": "system", "content": "You are a concise assistant." },
{ "role": "user", "content": "Explain inference in one paragraph." }
]
}'Python (OpenAI SDK compatible)
from openai import OpenAI
client = OpenAI(
api_key="YOUR_HUDDLE_API_KEY",
base_url="https://gru.huddle01.io/v1",
)
resp = client.chat.completions.create(
model="claude-sonnet-4",
messages=[
{"role": "user", "content": "Write a short release note for a new AI API."}
],
)
print(resp.choices[0].message.content)JavaScript (OpenAI SDK compatible)
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.HUDDLE_API_KEY,
baseURL: "https://gru.huddle01.io/v1",
});
const response = await client.chat.completions.create({
model: "deepseek-chat",
messages: [{ role: "user", content: "Give me 3 product tagline options." }],
});
console.log(response.choices[0].message?.content);Model switching example
Inference routing is model-driven. Keep the same code, change only model.
{ "model": "gpt-4o-mini" }
{ "model": "claude-sonnet-4" }
{ "model": "gemini-2.5-flash" }
{ "model": "deepseek-r1" }What “supported” means
Text/Chat: standard chat completion use cases.Code: generation and code transformation tasks.Reasoning: strong multi-step problem solving.Vision: multimodal image + text support where available.