10-16-2026, 03:13 PM
That matches what I was seeing. I don't think your conclusion is far off. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.
|
speculative decoding in llama.cpp on dual 3090
|
|
« Next Oldest | Next Newest »
|
| Messages In This Thread |
|
speculative decoding in llama.cpp on dual 3090 - by vectorport - 10-16-2026, 04:46 AM
Re: speculative decoding in llama.cpp on dual 3090 - by rasterbench - 10-16-2026, 06:27 AM
Re: speculative decoding in llama.cpp on dual 3090 - by pixelcrate82 - 10-16-2026, 03:13 PM
Re: speculative decoding in llama.cpp on dual 3090 - by visionforge - 10-18-2026, 03:41 AM
Re: speculative decoding in llama.cpp on dual 3090 - by mellowwire98 - 10-18-2026, 04:26 PM
Re: speculative decoding in llama.cpp on dual 3090 - by vrambandit - 10-18-2026, 06:49 PM
Re: speculative decoding in llama.cpp on dual 3090 - by mangotinker73 - 10-19-2026, 12:32 AM
|
Users browsing this thread: 1 Guest(s)