llama.cpp
graph : reduce splits for recurrent and hybrid models
#14825
Merged

graph : reduce splits for recurrent and hybrid models #14825

compilade
compilade graph : avoid creating redundant s_copy views
ac149794
compilade compilade added performance
gabe-l-hart
compilade
slaren
slaren
slaren approved these changes on 2025-07-23
gabe-l-hart
gabe-l-hart approved these changes on 2025-07-23
gabe-l-hart
compilade graph : comment the s_copy views
7b660a91
ggerganov
ggerganov approved these changes on 2025-07-24
gabe-l-hart
gabe-l-hart
ggerganov ggerganov merged 66625a59 into master 180 days ago
ggerganov ggerganov deleted the compilade/recurrent-reduce-splits branch 180 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone