Varity AI Gateway

Varity Team Core Contributors Updated August 2026

The Varity AI Gateway is an OpenAI-compatible inference API. It speaks the same chat-completions wire format as the OpenAI API, so every OpenAI SDK, framework adapter, and agent runtime already knows how to talk to it. The integration is a one-line base-URL swap.

https://varity.app/v1

Fastest Path

Set two variables and send a request. This is the whole integration.

curl

curl https://varity.app/v1/chat/completions \
  -H "Authorization: Bearer $VARITY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-5-2",
    "messages": [
      { "role": "user", "content": "Say hello in one sentence." }
    ]
  }'

Python

# pip install openai
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://varity.app/v1",
    api_key=os.environ["VARITY_API_KEY"],
)

response = client.chat.completions.create(
    model="glm-5-2",
    messages=[{"role": "user", "content": "Say hello in one sentence."}],
)

print(response.choices[0].message.content)

Node

// npm install openai
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://varity.app/v1",
  apiKey: process.env.VARITY_API_KEY,
});

const response = await client.chat.completions.create({
  model: "glm-5-2",
  messages: [{ role: "user", content: "Say hello in one sentence." }],
});

console.log(response.choices[0].message.content);

base_url / baseURL is the only line that differs from an OpenAI setup. Model ids come from the Varity catalog, not OpenAI’s. See Models for the full list.

Confirm The Gateway Before You Have A Key

GET /v1/models is public. Run it right now, with no credential, and you get a 200 plus the live catalog:

curl https://varity.app/v1/models

{
  "object": "list",
  "data": [
    {
      "id": "glm-5-2",
      "display_name": "GLM 5.2",
      "object": "model",
      "owned_by": "varity",
      "capabilities": {
        "chat_completions": true,
        "tool_calls": false,
        "streaming": true,
        "streaming_tool_calls": false
      },
      "context_window": 1000000,
      "max_completion_tokens": 131072,
      "privacy": "private",
      "privacy_routes": ["private"],
      "availability": { "status": "available" },
      "customer_pricing": {
        "status": "configured",
        "currency": "usd",
        "unit": "per_million_tokens",
        "input": 1.47,
        "output": 4.62,
        "cached_input": 1.47
      }
    }
  ]
}

The gateway also exposes an unauthenticated health probe:

curl https://varity.app/health

{ "status": "healthy", "service": "varity-gateway", "version": "1.13.49" }

Authentication

Inference requests require a Varity API key sent as a bearer credential:

Authorization: Bearer $VARITY_API_KEY

Create keys in the Developer Portal settings page. Key plaintext is shown once at creation and is never returned again. The same key format is used across the Varity public API.

Auth is enforced on /v1/chat/completions. A request with no Authorization header returns 401:

{
  "error": {
    "message": "A Bearer API key is required.",
    "type": "authentication_error",
    "code": "authentication_required",
    "param": null,
    "retryable": false,
    "request_id": "516c24ef-ba5a-4e3d-8f32-0a609ae5d298"
  }
}

A malformed or revoked key returns 401 with code: "invalid_api_key" and the message The API key is invalid or revoked.

Note

Authentication is checked before the request body is validated. An unauthenticated request naming a nonexistent model still returns the authentication error, not a model error.

Streaming

Set stream: true for server-sent events, exactly as with OpenAI. Every model in the catalog reports "streaming": true.

curl

curl https://varity.app/v1/chat/completions \
  -H "Authorization: Bearer $VARITY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-5-2",
    "stream": true,
    "messages": [
      { "role": "user", "content": "Count to five." }
    ]
  }'

Python

stream = client.chat.completions.create(
    model="glm-5-2",
    messages=[{"role": "user", "content": "Count to five."}],
    stream=True,
)

for chunk in stream:
    delta = chunk.choices[0].delta.content
    if delta:
        print(delta, end="")

Node

const stream = await client.chat.completions.create({
  model: "glm-5-2",
  messages: [{ role: "user", content: "Count to five." }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}

Streaming tool calls are a narrower capability than streaming text, and are only supported on some models. Check capabilities.streaming_tool_calls before you combine stream: true with tools. See Compatibility.

Errors

Errors use the OpenAI error envelope, with two Varity additions: retryable and request_id.

Field	Meaning
error.message	Human-readable description
error.type	Error class, for example authentication_error
error.code	Stable machine code, for example invalid_api_key
error.param	Offending request field, or null
error.retryable	Whether retrying the same request can succeed
error.request_id	Correlation id, also returned as the x-request-id response header

Quote request_id in any support request. It is present on both the response body and the x-request-id header on every response, including successful ones.

Rate Limits

GET /v1/models returns standard rate-limit headers:

ratelimit-limit: 120
ratelimit-policy: 120;w=60
ratelimit-remaining: 119
ratelimit-reset: 60

That is 120 requests per 60-second window on the catalog endpoint. Read the ratelimit-remaining and ratelimit-reset headers rather than hardcoding a budget; the policy is served live and can change.

Next Steps

• Models: the full live catalog with context windows, limits, and prices
• Compatibility: exactly what the OpenAI surface includes and excludes
• Public API Reference: the /api platform contract for deploys and operations