katanaml / sparrow

Data processing with ML, LLM and Vision LLM
https://katanaml.io
GNU General Public License v3.0
3.71k stars 378 forks source link

Sparrow parse model repository not found #73

Closed SalvadorRuizGuevara closed 1 month ago

SalvadorRuizGuevara commented 1 month ago

When im trying to use the sparrow-parse agent it gives the following error:

huggingface_hub.errors.RepositoryNotFoundError: 404 Client Error. (Request ID: Root=1-67004a09-066d8f903d8c88d071f3c213;b02687c7-fa52-4c92-8120-36baf4e06e1d)

Repository Not Found for url: https://huggingface.co/api/spaces/katanaml/sparrow-qwen2-vl-7b. Please make sure you specified the correct repo_id and repo_type. If you are trying to access a private or gated repo, make sure you are authenticated.

Is the sparrow parse model deprecated? Or moved to another repository?

abaranovskis-redsamurai commented 1 month ago

This is correct, this HF space is private, as it runs on paid GPU. You can create your own HF space and deploy the code available in Sparrow Parse: https://github.com/katanaml/sparrow/tree/main/sparrow-data/parse/sparrow_parse/vllm/infra/qwen2_vl_7b

mrtj commented 1 month ago

May I suggest to update the documentation with this info? I really struggle to get anything working in this repo, out of the unstructured wrapper. What is "sparrow key" for example?

abaranovskis-redsamurai commented 1 month ago

I'm in the process of updating and simplifying docs structure currently. In the next version agents and code related to Unstructured, Haystack, LlamaIndex and LangChain will be removed. As my primary focus on Sparrow Parse functionality with vision LLM models. Sparrow Parse generates the best results with vision LLM.

Sparrow key can be set to anything in your own deployment. I use Sparrow key to control access to Sparrow deployment on GPU instance, for example: https://katanaml-sparrow-ui.hf.space. I give Sparrow key only to potential customers.