Content
@
https://warpcast.com/~/channel/aichannel
0 reply
0 recast
0 reaction
caz.eth
@caz.eth
DeepSeek with R1 is a phenomenal free product as long as the servers are not busy. Anyone else use it a lot and have some hacks/tips?
1 reply
0 recast
0 reaction
engineer monster
@vicc
if you don’t want to spin up your own server to run the 671gb model, check out @venice-ai (staking VVV gives you a certain amount of free inference credits), otherwise you can run a distilled version locally (I’m running the 16b one and could probs get away with the 32b on my 96gb mac laptop)
1 reply
0 recast
1 reaction
caz.eth
@caz.eth
Is Venice good?
1 reply
0 recast
0 reaction
engineer monster
@vicc
I haven’t used it via API yet although I have some compute units as a staker, but it’s essentially a host for the top open source models including Deepseek 671B and Flux.
0 reply
0 recast
1 reaction