OpenAI

Setup

import carrot_ai
from openai import OpenAI

carrot_ai.init(api_key="sk-...")
client = carrot_ai.wrap(OpenAI())

Chat completions

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "What is the capital of France?"},
    ],
    temperature=0.7,
)

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

Streaming

Streaming works transparently — iterate as usual and the trace is captured when the stream finishes:

stream = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Tell me a story"}],
    stream=True,
)

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

Tool calls

Tool calls are captured automatically, including function names, arguments, and IDs:

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "What's the weather in SF?"}],
    tools=[{
        "type": "function",
        "function": {
            "name": "get_weather",
            "description": "Get current weather",
            "parameters": {
                "type": "object",
                "properties": {"city": {"type": "string"}},
            },
        },
    }],
)

Async client

from openai import AsyncOpenAI

client = carrot_ai.wrap(AsyncOpenAI())


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

With `@trace` for pipelines

Combine wrap() with @trace to capture multi-step workflows:

@carrot_ai.trace
def summarize_and_translate(text: str, language: str) -> str:
    summary = client.chat.completions.create(
        model="gpt-4o",
        messages=[
            {"role": "system", "content": "Summarize concisely."},
            {"role": "user", "content": text},
        ],
    ).choices[0].message.content

    translation = client.chat.completions.create(
        model="gpt-4o",
        messages=[
            {"role": "system", "content": f"Translate to {language}."},
            {"role": "user", "content": summary},
        ],
    ).choices[0].message.content

    return translation

Both LLM calls appear as child traces under the parent summarize_and_translate trace.

Default tags

Add default tags to every trace from a specific client:

client = carrot_ai.wrap(OpenAI(), tags=["production", "gpt-4o"])