Running asyncio.get_event_loop().run_until_complete(query(args)) more than once deadlocks program about lmql HOT 5 CLOSED

I think there may be an issue with worker threads being shut down after the first completion. I will investigate a bit and report here.

For now, usually when running multiple queries in sequence, I would suggest a pattern like this:

async def main():
   l = []
   await s(l, "some text")
   await s(l, "some text")

asyncio.run(main)

from lmql.

Thank you for the quick answer. sadly I am calling the functions from within a RabbitMq callback that I cannot make async and since I the function gets called for each message I need a way to transition more than once from non-async to async

from lmql.

I see. For this you may want to have a look at how we bridge async/non-async for langchain integration, see https://github.com/eth-sri/lmql/blob/main/src/lmql/runtime/langchain.py

Bascially, if possible try re-using the event loop. Still, I will see what I can do to enable your initial use.

from lmql.

Thanks a lot!! I am triying to see exactly where it gets locked to have a better idea on whats happening
so far it gets locked on
lmql/runtime/bopenai/batched_openai.py :675
PS I am using version 0.0.5.1

from lmql.

It worked thanks!!

from lmql.