nhbis0n

joined 1 year ago
sorted by: new top controversial old
llama2 13B on Gtx 1070 in c/localllama@poweruser.forum
[–] nhbis0n@alien.top 1 points 11 months ago

I run 7B’s on my 1070. ollama run llama2 produces between 20 and 30 tokens per second in ubuntu.

1
M3 Pro (alien.top)
 

Anyone know the largest model size that will fit into the new M3 Pro with 36GB RAM? I am looking to run some 23GB models with long context.

OpenAI announces leadership transition: Sama no longer CEO in c/localllama@poweruser.forum
[–] nhbis0n@alien.top 1 points 11 months ago

Did the AI coordinate his sacking?