Razvan Dinu

We don't have a patch release planned for this. We'll release 0.10.0 at the end of the month, and it will be part of that.

This has now been released as part of `0.9.1`. The `0.10.0` is pushed out for a few more weeks.

Hi @pechaut78! You are correct that the behavior is weird. This is because the "|" operator actually mutates the `RunnableRails` instance (https://github.com/NVIDIA/NeMo-Guardrails/blob/develop/nemoguardrails/integrations/langchain/runnable_rails.py#L86). This is a bug, it should create a...

We should be able to do some inspection of the pydantic model to figure out the right way to pass it. Let me record this tentatively for 0.10.0.

Hi @gingters! I'm sorry to hear that you're having so much trouble getting this to work. Unfortunately, streaming is not yet supported in the `RunnableRails` (https://github.com/NVIDIA/NeMo-Guardrails/blob/develop/docs/user_guides/langchain/runnable-rails.md#limitations). For the second issue,...

Indeed. We could even use "model" as the configuration id. Thanks for the contribution!

Thanks for reporting this @praveen062! Yes, sometimes the LLM can break the structure of the response. This can be fixed with additional stop words or with a better output parsing....

Thanks @pradeepdev-1995 and @sciftlikci! Configuring the threshold is scheduled for `0.10.0`. To improve this temporarily, you can add more examples to the on-topic messages.

Hi @piotrm0! Yes, we are aware of this issue. We'll target to fix this for the `0.9.0` release.

Hi @cyun9601! Is this what you're looking for: https://github.com/NVIDIA/NeMo-Guardrails/blob/main/docs/user_guides/server-guide.md#threads?