Ngoc Le Dinh

> > It is not exactly what you asked, but wouldn't 'agent.iter' work in this case? > > from pydantic_ai import CallToolsNode > > > > agent = Agent(...) >...

> What LLM model do you use? and what prompt you use to create the community report? I use `gpt-4-turbo-preview`. About the prompt first I tried prompt tunning for my...

> @Nuclear6 if you trying prompt tune, you should try a large language model and optimize for chinese like qwen or moonshot. Before I was using gemma2 9b, it is...

@KylinMountain Hey, I've tried this `python -m graphrag.prompt_tune --root . --domain "Chinese web novels" --language Chinese --chunk-size 300 --output prompt_zh` but it said there is no language argument? did u...

Hi @KylinMountain did you notice that some records in the community records are English? It makes the result is English too if there is a question use that records. Where...

Hello, have this issue been solved?

> Isn't this solved by setting the `parallel_tool_calls` to `False`? 🤔 > > I think this is solved... Yes you're right! Thank you: ### Reference: https://ai.pydantic.dev/api/settings/#pydantic_ai.settings.ModelSettings.parallel_tool_calls

hi @royalsanga24 does your script generate `expected_output` as I read somewhere they intentionally assign `expected_output=None` to prevent hallucination