Inference › Optimization & Performance › Context / KV Cache v
speculative decoding actually worth it with PyTorch

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Linear Mode
speculative decoding actually worth it with PyTorch
vrambandit Offline
Member
***
Posts: 168
Threads: 14
Joined: Sep 2026
Reputation: 0
#6
10-08-2026, 05:26 PM
Quote:Tried to reproduce it just now. My result was mixed: first run looked bad, second run looked normal. The surprise was that memory pressure showed up b

same here on my side.
2x3090 • llama.cpp • too much VRAM is still not enough
Find
Reply


Messages In This Thread
Re: speculative decoding actually worth it with PyTorch - by vrambandit - 10-08-2026, 05:26 PM

Forum Jump:


Users browsing this thread: 1 Guest(s)