acty-openai

acty-openai provides an OpenAIExecutor implementation for Acty. It adapts langchain-openai chat models to the Acty executor interface and emits the shared telemetry attributes used across the wider Acty stack.

Install

pip install acty-openai

For local development:

pip install -e .[dev]

Usage

import asyncio

from acty import ActyEngine, EngineConfig
from acty_openai import OpenAIExecutor
from langchain_openai import ChatOpenAI


async def main() -> None:
    model = ChatOpenAI(model="gpt-4o-mini")
    engine = ActyEngine(
        executor=OpenAIExecutor(model=model),
        config=EngineConfig(primer_workers=1, follower_workers=1),
    )
    try:
        payload = {
            "messages": [{"role": "user", "content": "hello"}],
        }
        submission = await engine.submit_group("demo", payload, [])
        if submission.primer is not None:
            result = await submission.primer
            print(result.output)
    finally:
        await engine.close()


asyncio.run(main())

Notes

• if you do not pass a model explicitly, OpenAIExecutor requires langchain-openai
• the executor can attach OpenTelemetry span attributes when telemetry is enabled
• this package depends directly on both acty and acty-core because it imports both at runtime

Development

• tests live under tests/
• the repo includes unit tests plus an Acty engine integration test for shared telemetry behavior