arnavsinghvi11

LGTM! tagging @isaacbmiller as it seems like the ruff fix workflow is erroring out. Is [this](https://github.com/stanfordnlp/dspy/actions/runs/8156255381/job/22307837624?pr=559#step:5:124) something on the workflow end? or does it need to be handled within the...

> If you run `ruff check --fix-only`, then commit, it will pass CI. @software-artisan

tagging @isaacbmiller on failed tests. similar issues to previous PRs - could be related to #569 ?

@pmenkidoo just following up on this. can you rebase+merge with main and push again? the PR should be good to merge after that!

Thanks @mrshu !

Hi @BlueKiji77 you can use [HFClientTGI](https://dspy-docs.vercel.app/docs/deep-dive/language_model_clients/local_models/HFClientTGI) to support HuggingFace models and configure this to be your lm through `dspy.settings(lm = ...`)

Hi @BlueKiji77 , you can try using [HFModel](https://github.com/stanfordnlp/dspy/blob/main/dsp/modules/hf.py) but this is not recommended. Feel free to push a PR to support HuggingFacePipeline in DSPy!

Hi @ujjawal-ti , Could you specify which teleprompter you are using to compile, and also provide the full stack trace?

This seems unrelated to DSPy and more so from the HuggingFace model loading incompatibility, but I would point you to trying TGI since HFModel is less user-friendly for fixing these...

Yup! Their documentation on [quantization](https://huggingface.co/docs/text-generation-inference/en/conceptual/quantization) should be helpful.