Federico Bianchi

Hello and thanks for this! We have an upcoming PR (#120) that introduces litellm as an engine. This should hopefully cover most models (bedrock too) with one single implementation! Let...

sounds good, can you make ``` langchain>=0.3.0 langchain-aws>=0.2.1 ``` optional like in [grok](https://github.com/zou-group/textgrad/blob/main/textgrad/engine/groq.py)? we can merge this after that

thanks again for this!

Hi! I'll add these details to the repo as soon as possible (and also to the paper since they are missing, thanks for pointing this out). For the tags, we...

This should be supported by the new engines! Let us know if they work as expected!

Hello! Support for open-source is already here! we have quite a few [engines](https://github.com/zou-group/textgrad/tree/main/textgrad/engine) that cover most of open-source models. We don't yet have a hugging-face-specific implementation. See also this [implementation](https://github.com/zou-group/textgrad/blob/main/textgrad/engine/local_model_openai_api.py)

hello thanks so much! we are working on some updates so we can't merge this yet but will do so soon!

Hello! Yea this is me using old version of the packages (in this case, sklearn vs scikit-learn). Quick solution is: `git clone https://github.com/vinid/quica` open the requirements file locally and change...

Yea the issue could be that. But I am surprised it does not load the model on CPU in that case. Here we load it as CPU if it cuda...

Can you share the snippet you are trying to use? Be sure that caching is not active. We have two engines an older more stable version that supports parameters and...