lans_throwaway

joined 11 months ago
sorted by: new top controversial old
1
Look ahead decoding offers massive (~1.5x) speedup for inference (lmsys.org)
1
Look ahead decoding offers massive (~1.5x) speedup for inference (lmsys.org)
Dynamic LoRAs -- Crazy idea? in c/localllama@poweruser.forum
[–] lans_throwaway@alien.top 1 points 11 months ago (1 children)

Isn't that just Hypernetwork? It's been done before, eg. for stable diffusion