INTEGRATIONS/
HuggingFace
Connect Skydive agents to your Hugging Face
Connect Hugging Face and your agent finds models, reads their cards, and runs inference so you get results fast. Your ML work moves quicker while you do the work that matters.
Connect HuggingFace
→Learn more
Hugging Face, connected
Turn your Skydive agents into ML teammates
One connection, and your agent searches the hub, reads model cards, and runs inference so you can test ideas fast.
Just ask, in plain English
Ask your agent for what you need, right where you already work. It handles the rest in HuggingFace.
Find a model
Model card
Run inference
#
models
James
9:18 AM
Find a good open model for summarization.
Coppery
APP
9:19 AM
Searched the hub for summarization. Three open models fit, and the top one is small enough to run cheaply, so I noted its size and license.
James
9:24 AM
Perfect, thanks.
#
datasets
Brendan
8:31 AM
What does this model card say about license and size?
Link
APP
8:32 AM
Read the model card. It is Apache 2.0 licensed, 7 billion parameters, and the card lists the training data, so I summarized the parts that matter.
Brendan
8:38 AM
Good to know.
#
inference
Eldar
9:00 AM
Run this text through a sentiment model.
Fleet
APP
9:02 AM
Ran your text through a sentiment model. It came back positive with high confidence, and I pulled the scores so you can see the breakdown.
Eldar
9:05 AM
Nice, helpful.
Live in three steps
No integration project. Connect in under a minute and your agent is working.
01
Find
Pick the integration from your agent's list, or just ask the agent to connect it.
02
Approve
Confirm the access. Change what's granted anytime.
03
Done
Your agent gets to work and reports back.
⇄
Connect your agent to Hugging Face today
Connect HuggingFace
→Browse all integrations