OpenAI SDK - Edgee documentation

Edgee provides an OpenAI-compatible API, which means you can use the official OpenAI SDKs for TypeScript and Python with Edgee. This allows you to leverage Edgee’s routing, observability, and cost tracking features without changing your existing code.

Why Use OpenAI SDK with Edgee?

Up to 50% Cost Reduction: Automatic token compression when enabled via headers or console
No Code Changes: Use your existing OpenAI SDK code as-is
Multi-Provider Access: Route to OpenAI, Anthropic, Google, and more through one API
Automatic Failover: Built-in reliability with fallback providers
Cost Tracking: Real-time visibility into token usage in the Edgee dashboard
Observability: Request tracing, compression metrics, and logging across all providers

Installation

Install the OpenAI SDK for your preferred language:

TypeScript
Python

npm install openai

pip install openai

Configuration

Configure the OpenAI SDK to use Edgee’s API endpoint:

import OpenAI from "openai";

const openai = new OpenAI({
  baseURL: "https://edgee.io/v1",
  apiKey: process.env.EDGEE_API_KEY, // Your Edgee API key
});

async function main() {
  const completion = await openai.chat.completions.create({
    model: "gpt-5.2",
    messages: [
      { role: "user", content: "What is the capital of France?" }
    ],
  });

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

main();

from openai import OpenAI
from os import getenv

# Configure OpenAI SDK to use Edgee
client = OpenAI(
  base_url="https://edgee.io/v1",
  api_key=getenv("EDGEE_API_KEY"),  # Your Edgee API key
)

completion = client.chat.completions.create(
  model="gpt-5.2",
  messages=[
    {
      "role": "user",
      "content": "What is the capital of France?",
    },
  ],
)

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

Token Usage Tracking

Access standard OpenAI token usage metrics in every response:

import OpenAI from "openai";

const openai = new OpenAI({
  baseURL: "https://edgee.io/v1",
  apiKey: process.env.EDGEE_API_KEY,
});

const completion = await openai.chat.completions.create({
  model: "gpt-5.2",
  messages: [
    { role: "user", content: "Summarize this long document..." }
  ],
});

console.log(completion.choices[0].message.content);
console.log(`Prompt tokens: ${completion.usage?.prompt_tokens}`);
console.log(`Completion tokens: ${completion.usage?.completion_tokens}`);
console.log(`Total tokens: ${completion.usage?.total_tokens}`);

from openai import OpenAI
from os import getenv

client = OpenAI(
  base_url="https://edgee.io/v1",
  api_key=getenv("EDGEE_API_KEY"),
)

completion = client.chat.completions.create(
  model="gpt-5.2",
  messages=[
    {"role": "user", "content": "Summarize this long document..."}
  ],
)

print(completion.choices[0].message.content)
print(f"Prompt tokens: {completion.usage.prompt_tokens}")
print(f"Completion tokens: {completion.usage.completion_tokens}")
print(f"Total tokens: {completion.usage.total_tokens}")

When compression is enabled, prompt_tokens reflects the compressed token count. View detailed compression metrics in the Edgee dashboard.

Compression & Tags via Headers

When using the OpenAI SDK with Edgee, you can control token compression and add tags using HTTP headers:

Enabling Compression

import OpenAI from "openai";

const openai = new OpenAI({
  baseURL: "https://edgee.io/v1",
  apiKey: process.env.EDGEE_API_KEY,
  defaultHeaders: {
    "x-edgee-compression-model": "claude",
    "x-edgee-tags": "production,openai-sdk,user-123",
  },
});

from openai import OpenAI
from os import getenv

client = OpenAI(
  base_url="https://edgee.io/v1",
  api_key=getenv("EDGEE_API_KEY"),
  default_headers={
    "x-edgee-compression-model": "claude",
    "x-edgee-tags": "production,openai-sdk,user-123",
  },
)

Available Headers:

Header	Type	Description
`x-edgee-compression-model`	`"claude"`, `"opencode"`, `"cursor"`, `"codex"`	Compression bundle to apply (e.g. “claude” for Claude Token Compression)
`x-edgee-tags`	`string`	Comma-separated tags for analytics and filtering

You can also enable compression per API key or Agent in the Edgee console. Headers override console settings for specific requests.

Advanced Usage

Function Calling (Tools)

Edgee fully supports OpenAI’s function calling interface:

import OpenAI from "openai";

const openai = new OpenAI({
  baseURL: "https://edgee.io/v1",
  apiKey: process.env.EDGEE_API_KEY,
});

async function main() {
  const completion = await openai.chat.completions.create({
    model: "gpt-5.2",
    messages: [
      { role: "user", content: "What's the weather in Paris?" }
    ],
    tools: [
      {
        type: "function",
        function: {
          name: "get_weather",
          description: "Get the current weather for a location",
          parameters: {
            type: "object",
            properties: {
              location: {
                type: "string",
                description: "City name",
              },
            },
            required: ["location"],
          },
        },
      },
    ],
    tool_choice: "auto",
  });

  if (completion.choices[0].message.tool_calls) {
    console.log("Tool calls:", completion.choices[0].message.tool_calls);
  } else {
    console.log("Response:", completion.choices[0].message.content);
  }
}

main();

from openai import OpenAI
from os import getenv

client = OpenAI(
  base_url="https://edgee.io/v1",
  api_key=getenv("EDGEE_API_KEY"),
)

completion = client.chat.completions.create(
  model="gpt-5.2",
  messages=[
    {
      "role": "user",
      "content": "What's the weather in Paris?",
    },
  ],
  tools=[
    {
      "type": "function",
      "function": {
        "name": "get_weather",
        "description": "Get the current weather for a location",
        "parameters": {
          "type": "object",
          "properties": {
            "location": {
              "type": "string",
              "description": "City name",
            },
          },
          "required": ["location"],
        },
      },
    },
  ],
  tool_choice="auto",
)

if completion.choices[0].message.tool_calls:
    print("Tool calls:", completion.choices[0].message.tool_calls)
else:
    print("Response:", completion.choices[0].message.content)

Streaming Responses

Edgee supports streaming responses for real-time token delivery:

import OpenAI from "openai";

const openai = new OpenAI({
  baseURL: "https://edgee.io/v1",
  apiKey: process.env.EDGEE_API_KEY,
});

async function main() {
  const stream = await openai.chat.completions.create({
    model: "gpt-5.2",
    messages: [
      { role: "user", content: "Tell me a short story" }
    ],
    stream: true,
  });

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

main();

from openai import OpenAI
from os import getenv

client = OpenAI(
  base_url="https://edgee.io/v1",
  api_key=getenv("EDGEE_API_KEY"),
)

stream = client.chat.completions.create(
  model="gpt-5.2",
  messages=[
    {
      "role": "user",
      "content": "Tell me a short story",
    },
  ],
  stream=True,
)

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

Migration from OpenAI

If you’re already using the OpenAI SDK, migrating to Edgee is straightforward:

Change the base URL: Update baseURL from https://api.openai.com/v1 to https://edgee.io/v1
Update API key: Use your Edgee API key instead of your OpenAI key
That’s it! Your existing code will work without any other changes

All OpenAI SDK features are supported, including streaming, function calling, and response formatting. Edgee maintains full compatibility with the OpenAI API specification.

​Why Use OpenAI SDK with Edgee?

​Installation

​Configuration

​Token Usage Tracking

​Compression & Tags via Headers

​Enabling Compression

​Advanced Usage

​Function Calling (Tools)

​Streaming Responses

​Migration from OpenAI

Why Use OpenAI SDK with Edgee?

Installation

Configuration

Token Usage Tracking

Compression & Tags via Headers

Enabling Compression

Advanced Usage

Function Calling (Tools)

Streaming Responses

Migration from OpenAI