Features

• No API key required — uses your browser session token
• Persistent sessions — messages stay in the same conversation automatically
• Streaming support — yield reply chunks in real time
• Web search — toggle Kimi's built-in search on any message
• Multi-turn conversations — full context preserved across calls
• Lightweight — only depends on requests

Install

pip install p2d-kimi

Quick Start

from p2dkimi import KimiClient

client = KimiClient(token="your-kimi-bearer-token")

reply = client.chat("What is the capital of France?")
print(reply)

Multi-Turn Conversation

Messages are automatically grouped into the same session. The client creates one session on the first call and reuses it for all subsequent calls.

client = KimiClient(token="your-token")

reply1 = client.chat("My name is Alex.")
reply2 = client.chat("What is my name?")
print(reply2)

Starting a New Session

Call new_chat() to explicitly start a fresh conversation.

client.new_chat(name="fresh-start")

reply = client.chat("Let's start over.")

Streaming

from p2dkimi import KimiClient

client = KimiClient(token="your-token")

for chunk in client.stream("Explain quantum computing in simple terms"):
    print(chunk, end="", flush=True)

With Web Search

reply = client.chat("Latest news on AI today", use_search=True)

Manual Chat ID

You can still pass a chat_id directly to target a specific conversation.

chat_id = client.new_chat(name="my-session")

reply1 = client.chat("My name is Alex.", chat_id=chat_id)
reply2 = client.chat("What is my name?", chat_id=chat_id)
print(reply2)

API Reference

KimiClient(token, model="kimi", timeout=60)

Methods

Getting Your Token

1. Open kimi.moonshot.cn in your browser
2. Open DevTools → Network → any request
3. Copy the value of Authorization: Bearer <token>

Or from iOS/Android using a MITM proxy (e.g. mitmproxy, Charles).

The token is a JWT that lasts ~30 days.

How It Works

KimiClient.chat("hello")
    → new_chat()          POST /api/chat            → chat_id (stored in self.chat_id)
    → stream_completion() POST /api/chat/{id}/completion/stream  → SSE chunks
    → collect chunks      event=cmpl text="Hello!"
    → return full string

KimiClient.chat("follow-up")
    → reuses self.chat_id (no new session created)
    → stream_completion() POST /api/chat/{id}/completion/stream  → SSE chunks

REST API Server

The included Flask server exposes HTTP endpoints:

Pass your token via Authorization: Bearer <token> header. Include chat_id in the body to continue an existing session.

Demo

See DEMO.md for a full working Telegram bot built with this library.

Community

Join for Python tutorials, projects, and updates → t.me/pythontodayz

License

MIT — made by addy