Anyone in #roamcult played with @huggingface yet.
Kinda biased but I think @wolfejosh tends to have good taste in startups
Kinda biased but I think @wolfejosh tends to have good taste in startups
https://twitter.com/wolfejosh/status/1369754805762154503
Also they look esoteric AF
https://twitter.com/asmesfer/status/1369548400002220034
The big question I have here is - If we want to allow a user to
A) Identify related blocks in their Roam graph (maybe using something like W2V)
B) Summarize a page or chapter they have imported
C) Get 4 additional suggested movies, based on a list of movies they entered
Etc.
A) Identify related blocks in their Roam graph (maybe using something like W2V)
B) Summarize a page or chapter they have imported
C) Get 4 additional suggested movies, based on a list of movies they entered
Etc.
Can we allow them to benefit from these language models without sharing any of their notes with ANY 3rd parties?
It seems like HuggingFace has enterprise option that might help us spin up our own infrastructure.
Any local/offline first case studies
Cc @wolfejosh @huggingface
It seems like HuggingFace has enterprise option that might help us spin up our own infrastructure.
Any local/offline first case studies
Cc @wolfejosh @huggingface
Ok seems possible
https://twitter.com/stasbekman/status/1368969281136779266
This is me acting a total noob in public, but looks like we can just straight up download the models, put them on our servers, and use them without any interaction w HuggingFace or the model creators?
https://twitter.com/AWSOpen/status/1367548863201112066
• • •
Missing some Tweet in this thread? You can try to
force a refresh
