|
speculative decoding in llama.cpp
|
10-11-2026, 04:18 PM
I'd approach it in a boring order. If you're on vLLM, I'd confirm the version first and then reduce variables until the behavior is obvious. Once you know the plain path works, then add the faster path back in piece by piece.
I'm happy to test one more variant if you want a second data point.
mostly local
|
|
« Next Oldest | Next Newest »
|
| Messages In This Thread |
|
speculative decoding in llama.cpp - by embershed99 - 10-07-2026, 01:45 AM
Re: speculative decoding in llama.cpp - by mangotinker73 - 10-09-2026, 02:08 AM
Re: speculative decoding in llama.cpp - by vrambandit - 10-09-2026, 10:48 AM
Re: speculative decoding in llama.cpp - by marbletrack - 10-09-2026, 07:04 PM
Re: speculative decoding in llama.cpp - by northcraft90 - 10-09-2026, 09:46 PM
Re: speculative decoding in llama.cpp - by vrambandit - 10-09-2026, 10:45 PM
Re: speculative decoding in llama.cpp - by visionyard23 - 10-10-2026, 01:23 AM
Re: speculative decoding in llama.cpp - by vrambandit - 10-10-2026, 02:52 AM
Re: speculative decoding in llama.cpp - by papermixer4 - 10-10-2026, 03:22 PM
Re: speculative decoding in llama.cpp - by sanepeak48 - 10-10-2026, 03:39 PM
Re: speculative decoding in llama.cpp - by macroforge - 10-11-2026, 04:18 PM
Re: speculative decoding in llama.cpp - by mangotinker73 - 10-12-2026, 12:59 AM
|
