Type to search
Advanced channel search

EdgeLLama

@edgellama

Hyperspace is working to make AI abundant.

• EdgeLLama
Peer to peer protocol to run a node and serve the community with inference and vectors.

• Collama.ai
Web app to curate and query Wikipedia-style datasets using RAG and millions of finetuned LLMs.

EdgeLLama

20 Jan, 22:47

https://x.com/varun_mathur/status/1823796915802108151
518
EdgeLLama

20 Jan, 20:36

https://x.com/hyperspaceai/status/1790881334472081918?s=46&t=4cXL3a06AzV7NDcbmvVVlg
1,225
EdgeLLama

20 Jan, 20:29

PS: it is time to update to v0.1 :) aios.network
1,457
EdgeLLama

20 Jan, 19:49

https://x.com/varun_mathur/status/1778876925647126567?s=46&t=4cXL3a06AzV7NDcbmvVVlg
1,782
EdgeLLama

20 Jan, 19:26

Just crossed 6000 nodes. Available for Mac, Windows and Linux now.

https://aios.network
2,112
EdgeLLama

20 Jan, 18:51

https://x.com/hyperspaceai/status/1764351806522671352?s=46&t=4cXL3a06AzV7NDcbmvVVlg
2,292
EdgeLLama

20 Jan, 18:50

We are close to 2000 nodes now. You can ask @internalhybot with the command /count to get an updated count anytime.

If you haven’t already installed, please do so to earn points. Major update + utilization of the node is coming up in April.

https://hyperspace.computer/node
2,136
EdgeLLama

20 Jan, 17:05

Simplified one-click setup, now available for both Intel and Silicon-based Macs. Serve Mistral-7B-Instruct-v0.1 to others using your spare compute.

https://hyperspace.computer/node
2,713
EdgeLLama

20 Jan, 16:58

EdgeLLama pinned «Setup isn’t straightforward right now but there are 61 nodes on the network already (these are M1/M2s serving a Mistral-7B finetune): https://hyperspace.computer/node»
0
EdgeLLama

20 Jan, 16:58

Setup isn’t straightforward right now but there are 61 nodes on the network already (these are M1/M2s serving a Mistral-7B finetune):

https://hyperspace.computer/node
2,317
EdgeLLama

20 Jan, 16:48

This is a full ChatGPT decentralized AI replacement which works today. You can click and choose "Community Node" on a repo in hyperspace.computer and the inference will be run on the P2P connected nodes and the result returned back to you. Has some bugs, but it's a start..
2,495
EdgeLLama

20 Jan, 16:47

EdgeLLama pinned «https://twitter.com/varun_mathur/status/1733267366568714456?»
0
EdgeLLama

20 Jan, 16:47

https://twitter.com/varun_mathur/status/1733267366568714456?
2,047
EdgeLLama

20 Jan, 16:35

EdgeLLama will be rolling out starting Dec 7th. Thank you for your patience. If you are in SF, join us here for a quick meet and greet. We will be hosting chats on Twitter Spaces as well for those who can’t make it in person.

https://lu.ma/rmbabchv
2,067
EdgeLLama

20 Jan, 16:20

Here is our “BitTorrent for AI” paper outlining the design further: https://twitter.com/varun_mathur/status/1726381112090591594?s=46&t=NvcypMf3Hn8aFb3mBnMjsw
2,054
EdgeLLama

20 Jan, 15:56

Marc Andreessen just reposted the EdgeLLama video - check it out.

https://twitter.com/varun_mathur/status/1720205483678609549?s=46&t=EK6sGjLkZ703MMiZd6rSnA
2,048
EdgeLLama

20 Jan, 15:45

EdgeLLama pinned «Hello, World!»
0
EdgeLLama

20 Jan, 15:45

Hello, World!
1,881
EdgeLLama

20 Jan, 15:45

Channel photo updated
0

2024 All Right Reserved