Quatarly API

Overview

Quatarly provides a unified REST API that proxies requests to Anthropic (Claude), Google (Gemini), and OpenAI (GPT) through a single endpoint and a single API key. All responses conform to the OpenAI Chat Completions schema, making it a true drop-in replacement anywhere you currently use api.openai.com.

Authentication

Every request must include your Quatarly API key as a Bearer token in the Authorization header.

Authorization: Bearer your-api-key-here

Get your free key at api.quatarly.cloud/api-key.

Base URL

All endpoints live under /v1. The difference is just what base URL you give each client — the client appends the rest automatically.

Rate Limits

Rate limits are enforced per API key. Limits vary by key tier and model. A 429 Too Many Requests response is returned when exceeded.

All Models

All models below are accessible via the standard OpenAI Chat Completions API format using your Quatarly key.

Claude (Anthropic)

Gemini (Google)

GPT (OpenAI)

Factory AI Droid

Connect Factory AI Droid to Quatarly to access all models with a single API key. The setup script patches ~/.factory/settings.json automatically with all model entries.

• Install Factory AI

  Windows macOS / Linux

  powershellCopy

  irm https://app.factory.ai/cli/windows | iex

  bashCopy

  curl -fsSL https://app.factory.ai/cli | sh

• Create a Factory Account & Login

  Go to app.factory.ai and create a free account. Then run droid in your terminal and log in to create ~/.factory/settings.json.

• Run the Setup Script

  You need your Quatarly API key (qua_trail_... or qua_...).

  Windows macOS / Linux

  powershellCopy

  irm https://raw.githubusercontent.com/himanshu91081/Quatarly-setup/main/add-quatarly-models.ps1 -OutFile add-quatarly-models.ps1; .\add-quatarly-models.ps1

  bashCopy

  curl -fsSL https://raw.githubusercontent.com/himanshu91081/Quatarly-setup/main/add-quatarly-models.sh -o add-quatarly-models.sh && bash add-quatarly-models.sh

  The script will prompt for your key and update settings.json with all 11 model entries. Running it again with a new key safely updates existing entries without creating duplicates.

Verify Setup

Select-String "customModels" -Path "$env:USERPROFILE\.factory\settings.json" -A 5

grep -A 5 "customModels" ~/.factory/settings.json

Expected snippet in settings.json:

"customModels": [
  {
    "model":       "claude-sonnet-4-6-thinking",
    "id":          "custom:claude-sonnet-4-6-thinking-0",
    "baseUrl":     "https://api.quatarly.cloud/",
    "apiKey":      "your-api-key",
    "provider":    "anthropic",
    "displayName": "claude-sonnet-4-6-thinking"
  },
  {
    "model":       "gpt-5.1",
    "id":          "custom:gpt-5.1-5",
    "baseUrl":     "https://api.quatarly.cloud/v1",
    "apiKey":      "your-api-key",
    "provider":    "openai",
    "displayName": "gpt-5.1"
  }
]

Claude Code

Use Claude Code as a CLI coding agent routed through Quatarly. No Anthropic account needed — just your Quatarly key.

• Install Claude Code

  bashCopy

  npm install -g @anthropic-ai/claude-code

• Set Environment Variables

  Option A — Setup script (recommended, persists across restarts):

  Windows macOS / Linux

  powershellCopy

  irm https://raw.githubusercontent.com/himanshu91081/Quatarly-setup/main/set-claude-env.ps1 -OutFile set-claude-env.ps1; .\set-claude-env.ps1 -ApiKey "your-api-key-here"

  bashCopy

  curl -fsSL https://raw.githubusercontent.com/himanshu91081/Quatarly-setup/main/set-claude-env.sh -o set-claude-env.sh && bash set-claude-env.sh your-api-key-here

  Option B — Set manually for current session only:

  macOS / Linux Windows

  bashCopy

  export ANTHROPIC_BASE_URL="https://api.quatarly.cloud/"
  export ANTHROPIC_AUTH_TOKEN="your-api-key-here"
  export ANTHROPIC_DEFAULT_HAIKU_MODEL="claude-haiku-4-5-20251001"
  export ANTHROPIC_DEFAULT_SONNET_MODEL="claude-sonnet-4-6-thinking"
  export ANTHROPIC_DEFAULT_OPUS_MODEL="claude-opus-4-6-thinking"

  powershellCopy

  $env:ANTHROPIC_BASE_URL             = "https://api.quatarly.cloud/"
  $env:ANTHROPIC_AUTH_TOKEN           = "your-api-key-here"
  $env:ANTHROPIC_DEFAULT_HAIKU_MODEL  = "claude-haiku-4-5-20251001"
  $env:ANTHROPIC_DEFAULT_SONNET_MODEL = "claude-sonnet-4-6-thinking"
  $env:ANTHROPIC_DEFAULT_OPUS_MODEL   = "claude-opus-4-6-thinking"

  Variable	Value
  ANTHROPIC_BASE_URL	https://api.quatarly.cloud/
  ANTHROPIC_AUTH_TOKEN	Your Quatarly API key
  ANTHROPIC_DEFAULT_HAIKU_MODEL	claude-haiku-4-5-20251001
  ANTHROPIC_DEFAULT_SONNET_MODEL	claude-sonnet-4-6-thinking
  ANTHROPIC_DEFAULT_OPUS_MODEL	claude-opus-4-6-thinking

  After using the setup script, run source ~/.zshrc (macOS) or source ~/.bashrc (Linux) to pick up the changes in the current terminal. GUI apps require a full logout/restart.
  Watch: Claude Code Setup Guide

• Launch Claude Code

  bashCopy

  claude

  Claude Code will route all requests through Quatarly using your key and credit balance.

OpenCode

Use OpenCode as a terminal AI coding assistant routed through Quatarly. No OpenAI account needed — just your Quatarly key.

• Install OpenCode

  bashCopy

  npm install -g opencode-ai

• Create the Config File

  Edit or create ~/.config/opencode/opencode.json:

  json (~/.config/opencode/opencode.json)Copy

  {
      "$schema": "https://opencode.ai/config.json",
      "provider": {
          "openai": {
              "options": {
                  "baseURL": "https://api.quatarly.cloud/v1",
                  "apiKey":  "your-api-key-here"
              }
          }
      },
      "model": "openai/gpt-5.3-codex"
  }

  Create it automatically from the terminal:

  Windows macOS / Linux

  powershellCopy

  $dir = "$env:USERPROFILE\.config\opencode"
  if (!(Test-Path $dir)) { New-Item -ItemType Directory -Force -Path $dir }
  @'
  {
      "$schema": "https://opencode.ai/config.json",
      "provider": {
          "openai": {
              "options": {
                  "baseURL": "https://api.quatarly.cloud/v1",
                  "apiKey": "your-api-key-here"
              }
          }
      },
      "model": "openai/gpt-5.3-codex"
  }
  '@ | Set-Content "$dir\opencode.json"

  bashCopy

  mkdir -p ~/.config/opencode
  cat > ~/.config/opencode/opencode.json << 'EOF'
  {
      "$schema": "https://opencode.ai/config.json",
      "provider": {
          "openai": {
              "options": {
                  "baseURL": "https://api.quatarly.cloud/v1",
                  "apiKey": "qua_trail_your-key-here"
              }
          }
      },
      "model": "openai/gpt-5.3-codex"
  }
  EOF

• Launch OpenCode

  bashCopy

  opencode

  OpenCode routes all requests through Quatarly. Switch models with /model inside the session — all Quatarly GPT, Gemini, and Claude models appear under the openai provider.

Chat Completions

Send a chat message to any model. The request and response format is identical to the OpenAI Chat Completions API.

Request Body

Example Request

{
  "model": "claude-sonnet-4-6-thinking",
  "messages": [
    { "role": "system", "content": "You are a helpful assistant." },
    { "role": "user",   "content": "Explain quantum entanglement simply." }
  ],
  "stream": false,
  "max_tokens": 1024
}

Response Format

{
  "id": "chatcmpl-xyz123",
  "object": "chat.completion",
  "created": 1750000000,
  "model": "claude-sonnet-4-6-thinking",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Quantum entanglement is when two particles..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 32,
    "completion_tokens": 118,
    "total_tokens": 150
  }
}

Parameters Reference

Common parameters across all models. Provider-specific parameters are passed through transparently.

cURL Examples

Basic Chat

curl -X POST "https://api.quatarly.cloud/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer your-api-key-here" \
  -d '{
    "model": "gpt-5.3-codex",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Streaming Response

curl -X POST "https://api.quatarly.cloud/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer your-api-key-here" \
  -N \
  -d '{
    "model": "claude-sonnet-4-6-thinking",
    "stream": true,
    "messages": [{"role": "user", "content": "Write a poem about the ocean."}]
  }'

Claude with System Prompt

curl -X POST "https://api.quatarly.cloud/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer your-api-key-here" \
  -d '{
    "model": "claude-opus-4-6-thinking",
    "messages": [
      {"role": "system", "content": "You are an expert Python developer."},
      {"role": "user",   "content": "Refactor this function for readability."}
    ],
    "max_tokens": 2048
  }'

Gemini with High Temperature

curl -X POST "https://api.quatarly.cloud/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer your-api-key-here" \
  -d '{
    "model": "gemini-3.1-pro",
    "messages": [{"role": "user", "content": "Brainstorm 10 startup ideas."}],
    "temperature": 1.4,
    "max_tokens": 1024
  }'

Claude via /v1/messages (Anthropic format)

If /v1/chat/completions isn't working with a Claude-only tool, try the Anthropic Messages endpoint instead. Note max_tokens is required and the system prompt is a top-level field.

curl -X POST "https://api.quatarly.cloud/v1/messages" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer your-api-key-here" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "claude-sonnet-4-6-thinking",
    "max_tokens": 1024,
    "system": "You are a helpful assistant.",
    "messages": [
      {"role": "user", "content": "Explain transformers in ML simply."}
    ]
  }'

Python (OpenAI SDK)

Install the official OpenAI library and point it at Quatarly — no other changes needed.

pip install openai

Basic Usage

from openai import OpenAI

client = OpenAI(
    api_key="your-api-key-here",
    base_url="https://api.quatarly.cloud/v1",
)

response = client.chat.completions.create(
    model="claude-sonnet-4-6-thinking",
    messages=[
        {"role": "user", "content": "Summarise the last decade of AI progress."}
    ],
    max_tokens=1024,
)

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

Streaming

with client.chat.completions.stream(
    model="gpt-5.3-codex",
    messages=[{"role": "user", "content": "Write a sorting algorithm."}],
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

Node.js (OpenAI SDK)

npm install openai

Basic Usage

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "your-api-key-here",
  baseURL: "https://api.quatarly.cloud/v1",
});

const response = await client.chat.completions.create({
  model: "gemini-3.1-pro",
  messages: [{ role: "user", content: "What is the capital of France?" }],
});

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

Streaming

const stream = await client.chat.completions.create({
  model: "claude-haiku-4-5-20251001",
  messages: [{ role: "user", content: "Tell me a short story." }],
  stream: true,
});

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

Models Table

Error Codes

All errors follow the OpenAI error response format with an additional code field.

{
  "error": {
    "message": "Rate limit exceeded. Please retry after 60 seconds.",
    "type":    "rate_limit_error",
    "code":    "rate_limit_exceeded"
  }
}

Script Files