Torrix Python SDK

Torrix is a self-hosted LLM observability platform. This SDK wraps your OpenAI and Anthropic clients to log every call automatically. Tokens, cost, latency, and full prompt traces go straight to your dashboard.

Installation

pip install torrix

Quick start

import torrix
from openai import OpenAI

torrix.init(api_key="trxk_...", base_url="http://localhost:8088")
client = torrix.wrap(OpenAI(api_key="sk-..."))

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

Anthropic

from anthropic import Anthropic

client = torrix.wrap(Anthropic(api_key="sk-ant-..."))

response = client.messages.create(
    model="claude-3-5-sonnet-20241022",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.content[0].text)

LangChain

from torrix.wrappers.langchain_callback import TorrixCallbackHandler
from langchain_openai import ChatOpenAI

torrix.init(api_key="trxk_...", base_url="http://localhost:8088")
handler = TorrixCallbackHandler()

llm = ChatOpenAI(model="gpt-4o-mini", callbacks=[handler])
response = llm.invoke("What is the capital of France?")

Optional extras

pip install torrix[openai]      # pins openai>=1.0
pip install torrix[anthropic]   # pins anthropic>=0.20
pip install torrix[all]         # both

Running Torrix

curl -o docker-compose.yml https://raw.githubusercontent.com/torrix-ai/install/main/docker-compose.community.yml
docker compose up

Opens at http://localhost:8088. Community edition is free forever.

Links

• Documentation
• Website
• Contact