Use Helix from anywhere.

Helix supports the OpenAI Chat Completions and Responses APIs, so compatible tools can use it from your terminal, editor, the official SDKs, or a self-hosted chat app. Chat Completions are stateless. Responses can be stored only when you explicitly set store: true.

1 · Get your key

Create an account and pick a plan.
Open Settings → API and create a key.
Copy it immediately. It starts with hx_ and is shown only once.

2 · The essentials

Base URL:   https://helixapp.net/v1
Auth:       Authorization: Bearer hx_your_key
Endpoints:  POST /v1/chat/completions
            POST /v1/responses
            GET  /v1/models

helix-auto	Recommended. Helix reads each request and applies the best orchestration automatically, including its deepest reasoning when the task calls for it.
helix-lite	Efficient non-frontier models for everyday work.
helix-core	Frontier models for maximum capability.
helix-ensemble	Always use ensemble reasoning: several independent lines of thought, synthesized into one answer.

3 · From the terminal

curl https://helixapp.net/v1/chat/completions \
  -H "Authorization: Bearer hx_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "helix-auto",
    "messages": [{"role": "user", "content": "Plan my product launch"}]
  }'

The response is standard OpenAI format: your answer is at choices[0].message.content.

4 · From code

Python, with the official OpenAI SDK:

from openai import OpenAI

client = OpenAI(base_url="https://helixapp.net/v1", api_key="hx_your_key")
r = client.chat.completions.create(
    model="helix-auto",
    messages=[{"role": "user", "content": "Review this contract clause…"}])
print(r.choices[0].message.content)

JavaScript / TypeScript:

import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://helixapp.net/v1", apiKey: "hx_your_key" });
const r = await client.chat.completions.create({
  model: "helix-auto",
  messages: [{ role: "user", content: "Summarize this report…" }],
});
console.log(r.choices[0].message.content);

5 · From chat apps you run yourself

Open WebUI: Admin Settings → Connections → add an OpenAI API connection with the base URL and your key. The four Helix modes appear in the model picker.

LibreChat: add a custom OpenAI compatible endpoint in librechat.yaml with the same base URL and key.

Anything else: clients that rely on Chat Completions or core Responses features can use the same base URL. Streaming, function tools, structured JSON, stored response retrieval, input-item listing and response deletion are supported.

Why developers choose this route

Chat Completions are processed without saving conversation content. Responses are also ephemeral by default; set store: true only when you want retrieval or continuation by response ID. Usage accounting still applies to both routes.

Multi turn conversations: include the previous messages in the messages array each time, exactly like the OpenAI API. Your history lives with you, not with us.