Content
@
0 reply
0 recast
0 reaction
Justin Hunter
@polluterofminds
What are the implications on on device models and how does that change the economics of AI? https://9to5google.com/2024/03/07/pixel-8-gemini-nano/
5 replies
0 recast
2 reactions
ππͺπΎπ‘π‘πΎ
@gm8xx8
side note android offers several benefits for deploying effective agents: - cloud-based emulators are straightforward to initiate. - the mobile view hierarchy is simpler than webpages, facilitating quicker, easier navigation for llms. - the ability to install apps on emulators enhances task completion capabilities
0 reply
0 recast
1 reaction
π _π£π _π
@m-j-r
I wonder if mobile models make for better compression. in the same vein, I wonder what the minimum requirements are for critical feedback to a world model. this may not just be inference cost trending to zero, but also valuable crowdsourcing of incremental gains (that can be incentivized).
1 reply
0 recast
1 reaction
MetaEndπ©
@metaend.eth
soo I will skip the 8 and probably go with the next model then
1 reply
0 recast
1 reaction
Siddharth Ramakrishnan
@siddharthvader
Makes marginal cost of inference go to 0. Makes AI more like traditional software from that perspective if it can be achieved
0 reply
0 recast
1 reaction
Chris Smith
@metapriest
It changes the trust model quite a bit. If I can run a model locally, it doesn't need to talk to server to ask permission to discuss something. Can also fine tune it on user responses, making it more personalized
0 reply
0 recast
1 reaction
