DustGrouchy1792

joined 10 months ago
sorted by: new top controversial old
Any tricks to speed up 13B models on a 3090? in c/localllama@poweruser.forum
[โ€“] DustGrouchy1792@alien.top 1 points 10 months ago (1 children)

Can I get koboldcpp working with sillytavern without too much of a headache?

Any tricks to speed up 13B models on a 3090? in c/localllama@poweruser.forum
[โ€“] DustGrouchy1792@alien.top 1 points 10 months ago

I'm now using a 4bit GPTQ version of the same model. After generation completes the VRAM goes up to 16.2 GB (out of 24 GB) and I have nothing else using GPU as best I can tell (no browser windows with youtube, etc).

Still only getting a bit under 4.00 tokens per second. So I don't think stuff is getting offloaded to CPU.

1
Any tricks to speed up 13B models on a 3090? (alien.top)
 

Are there any tricks to speed up 13B models on a 3090?

Currently using the regular huggingface model quantized to 8bit by a GPTQ capable fork of KoboldAI.

Especially when the context limit changes, it's pretty slow and far from even remotely real time.