Playback speed
×
Share post
Share post at current time
0:00
/
0:00

Build LLM Apps For LangFlow In HuggingFace Spaces

Via HuggingFace Spaces you have instant access to a fully fledged working version of the LangFlow IDE to build Generative Apps.

Thanks for reading Cobus Greyling on NLU, NLP, chatbots & voicebots! Subscribe for free to receive new posts and support my work.

The easiest way to access ⛓️LangFlow is via the Logspace space on HuggingFace. If you click on this url, the LangFlow IDE opens up, as seen below.

https://huggingface.co/Logspace

This environment is really reactive and the development components on the left-hand pane seem to be more than when I install LangFlow on replit.

Something I miss from LangFlow is the ability to search the components, in order to find a specific component. This feature is available in Flowise.

Flowise also have a web browser component which allows free web searches; as apposed to the Serp API which comes at a cost.

The workspace has tabs, so multiple flows can be open simultaneously and users can jump between tabs. Tab screens cannot be split in a such a way where multiple flows are displayed side-by-side.

As seen below, LangFlow also has a group of examples, analogous to Flowise’s marketplaces.

When chatting to the LangFlow application, the web socket connection URL used is: wss://logspace-langflow.hf.space/chat/ .

This socket was closed at some stage, but as can be seen in the image below, that problem has been fixed.

When clicking on the dialog button at the bottom-right, a window pops-up which allows you to interact with your application.

In an upcoming post I would love to do a wider comparison of LLM Application IDE’s.

Thanks for reading Cobus Greyling on NLU, NLP, chatbots & voicebots! Subscribe for free to receive new posts and support my work.

I’m currently the Chief Evangelist @ HumanFirst. I explore & write about all things at the intersection of AI and language; ranging from LLMs, Chatbots, Voicebots, Development Frameworks, Data-Centric latent spaces & more.

Cobus Greyling on LLMs, NLU, NLP, chatbots & voicebots
Authors
Cobus Greyling