Loader logic for LLaMA models
city96pushed 2 commits to main • 0b65261…8d1b56c • yesterday
city96pushed 1 commit to main • 65a7c89…0b65261 • 4 days ago
city96pushed 1 commit to main • 23b9817…65a7c89 • 8 days ago
city96pushed 1 commit to main • 8515647…23b9817 • 14 days ago
city96pushed 1 commit to main • 39102df…8515647 • 17 days ago
Stop custom tensor class propagating past dequant
city96pushed 2 commits to main • a8b5f5d…39102df • 17 days ago
city96pushed 1 commit to main • 4a84328…a8b5f5d • 18 days ago
city96pushed 1 commit to main • 57562d5…4a84328 • 28 days ago
city96pushed 1 commit to main • 51c9713…57562d5 • 29 days ago
Fix sd1 not using shape fix
city96pushed 1 commit to main • 8e898fa…51c9713 • on Nov 21
Add mochi to CLIPLoaderGGUF options (
#154 )
Pull request merge
city96pushed 1 commit to main • 9833348…8e898fa • on Nov 6
Fix for text encoder folder name change
city96pushed 1 commit to main • 6561064…9833348 • on Nov 3
city96pushed 1 commit to main • b702405…6561064 • on Oct 23
Fix SD3 patch metadata/tensor issue
city96pushed 1 commit to main • 4d7bb93…b702405 • on Oct 23
Add initial SD3 lcpp patch + instruction
city96pushed 1 commit to main • 0ef158d…4d7bb93 • on Oct 22
Add SD3 to loader arch list
city96pushed 1 commit to main • 68ad5fb…0ef158d • on Oct 22
city96pushed 1 commit to main • d090a91…68ad5fb • on Oct 15
Rewrite convert.py to work properly with dev de-distill
city96pushed 1 commit to main • d2aaeb0…d090a91 • on Oct 15
Always init tenc on offload device
city96pushed 1 commit to main • 454955e…d2aaeb0 • on Sep 17
Skip initialization for nn.Linear
city96pushed 1 commit to main • ac96699…454955e • on Sep 8
city96pushed 1 commit to main • 0342f0a…ac96699 • on Sep 6
city96pushed 1 commit to main • b46c206…0342f0a • on Sep 6
double patch issue fix for lowvram
city96pushed 1 commit to main • 717a0e1…b46c206 • on Sep 5
Temporarily disable instant patch logic
city96pushed 1 commit to main • c8923a4…717a0e1 • on Sep 3
mmap unload fix for lowvram
city96pushed 1 commit to main • 6dbb4ba…c8923a4 • on Sep 3
Temp fix for disable smart memory
city96pushed 1 commit to main • 69f0daf…6dbb4ba • on Sep 3
Merge pull request
#92 from city96/ops_refactor
Pull request merge
city96pushed 24 commits to main • 7f3ced6…69f0daf • on Sep 3
You can’t perform that action at this time.