Windows Copilot API: a free LLM API powered by Microsoft Copilot

Using your own Microsoft Copilot account. No API key, no credits, no paid plan: it turns the free chat at copilot.microsoft.com into an API you can call from code.

You can use it in two ways:

• 🐍 As a Python library: just call client.chat("Hi"). Supports streaming and multi-turn conversations.
• 🔌 As a local OpenAI-compatible API: runs a server at http://localhost:8000/v1 that speaks the OpenAI format, so the official openai SDK (and any OpenAI-compatible app) works as a drop-in, with localhost in place of OpenAI.

You sign in once with your Microsoft account in a browser; your session is saved and refreshed automatically after that.

Unofficial project. Not affiliated with or endorsed by Microsoft. It automates the consumer Copilot web experience for personal use, so use it responsibly and within Microsoft's terms.

---

Why use this?

• Free: uses your normal signed-in Copilot, no API billing.
• Drop-in OpenAI replacement: point any OpenAI client at localhost and it just works.
• Works everywhere you're signed in: the signed-in path works even in regions where anonymous Copilot is blocked (e.g. India).
• Streaming + conversations: token-by-token output and multi-turn threads addressed by conversation_id.

---

Requirements

• Python 3.9+
• A Microsoft account (the free one you use for Copilot is fine)
• Works on Windows, macOS, and Linux

---

Setup (2 minutes)

# 1. Clone the project
git clone <your-repo-url>
cd Windows-Copilot-API

2. Create and activate a virtual environment

On macOS / Linux:

python3 -m venv venv
source venv/bin/activate

On Windows (PowerShell):

python -m venv venv
venv\Scripts\Activate.ps1

On Windows you may need to allow script execution once: Set-ExecutionPolicy -Scope CurrentUser RemoteSigned. In cmd.exe activate with venv\Scripts\activate.bat instead.

3. Install dependencies and sign in

# Install dependencies
pip install -r requirements.txt

# Install the browser Playwright needs (one-time)
playwright install chromium

# Sign in once: a browser opens, log into your Microsoft account
python -m copilot login

That's it. Your session is saved under session/ (git-ignored, never shared) and reused on every run.

💡 You can even skip step 4: the first time you call chat() or start the server, it opens the sign-in browser for you automatically.

---

Usage 1: In Python (no server)

The simplest way if your code is already Python.

from copilot import CopilotClient

client = CopilotClient()                 # loads your signed-in session

# Get a full reply
reply = client.chat("Say hello in one short sentence.")
print(reply.text)

# Continue the SAME conversation — pass the id back
reply2 = client.chat("And now in French?", reply.conversation_id)
print(reply2.text)

# Stream the answer as it's typed
for chunk in client.stream("Tell me a short joke"):
    print(chunk, end="", flush=True)

chat() returns the full text plus a conversation_id; pass that id back to keep the thread going, or omit it to start fresh. stream() yields the reply piece by piece.

👉 More: examples/01_direct_chat.py, 02_direct_conversation.py, 03_direct_stream.py

---

Usage 2: As an OpenAI-compatible server

Start a local server that speaks the OpenAI API, so existing OpenAI tools and SDKs work unchanged.

python app.py
# -> Copilot OpenAI-compatible API on http://127.0.0.1:8000

Then point any OpenAI client at it (the API key is required by the SDK but ignored):

from openai import OpenAI

client = OpenAI(base_url="http://localhost:8000/v1", api_key="unused")

resp = client.chat.completions.create(
    model="copilot",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)

Or call it with plain HTTP / curl:

curl http://localhost:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"messages": [{"role": "user", "content": "Hello!"}]}'

Endpoints

Method	Path	Description
POST	/v1/chat/completions	Chat (supports "stream": true and an optional "conversation_id")
GET	/v1/models	Lists the single copilot model

Change the address with env vars: HOST=0.0.0.0 PORT=8080 python app.py, or run uvicorn server.api:app --host 0.0.0.0 --port 8080.

👉 More: examples/04_server_http.py, 05_server_stream.py, 06_server_openai_sdk.py

---

Command line

python -m copilot login          # sign in and save the session
python -m copilot ask "Hello!"   # quick one-shot question

---

Concurrency & stress test

The server bridges a single signed-in Copilot account, and Copilot's chat socket doesn't tolerate concurrent conversations from one process. So the server serializes upstream calls: parallel HTTP requests queue behind a lock and run one at a time (see server/api.py). This is intentional, and it means throughput is sequential, not parallel.

You can measure where it breaks with the included stress test, which fires a batch of simultaneous requests and doubles the batch size every successful round until the first error:

# Start the server in one terminal
python app.py

# Ramp concurrency in another (1 → 2 → 4 → 8 → …)
python tests/stress.py
python tests/stress.py --max 64 --timeout 120 --url http://localhost:8000

Sample run (one signed-in account):

Concurrency	Result	Wall time	Latency (min / median / max)
1	✓ all ok	3.7s	3.7 / 3.7 / 3.7s
2	✓ all ok	4.6s	3.4 / 4.6 / 4.6s
4	✓ all ok	8.3s	3.7 / 6.7 / 8.3s
8	✗ 1 failed (HTTP 502)	13.3s	3.5 / 9.7 / 13.3s

Highest fully-successful concurrency: 4. Wall time roughly doubles each round while minimum latency stays flat (~3.5s) — the signature of a serialized queue: one request runs immediately, the rest wait their turn. The failure at 8 is an upstream 502 (Copilot rejecting requests under load), not a server crash or timeout — so the exact break point is flaky and may vary between runs.

Takeaway: keep concurrent in-flight requests low (≈ 1–4). This is a personal bridge, not a high-throughput gateway — and please don't hammer your account.

---

Project layout

Path	What it does
copilot/	The core library: CopilotClient, auth, browser sign-in, HTTP driver
server/	The FastAPI OpenAI-compatible server
examples/	Runnable examples for every feature (examples/README.md)
tests/	Test scripts, including the concurrency stress test (tests/stress.py)
app.py	Starts the server

---

Notes & limitations

• Sign in once, then reuse. The cached token refreshes automatically; you only re-sign-in if the session fully expires.
• No daily limit, but be reasonable. Microsoft doesn't impose a daily chat cap, but please use it in moderation, and don't spam or hammer it with automated bulk requests.
• One model. Copilot has no model picker, so the server advertises a single model named copilot.
• Your session is private. Everything in session/ (cookies + token) stays on your machine and is git-ignored.

---

License

For personal and educational use. You are responsible for complying with Microsoft's terms of service.