summaryrefslogtreecommitdiff
path: root/candle-flash-attn
Commit message (Expand)AuthorAgeFilesLines
* Add some group parameter to convolutions. (#566)Laurent Mazare2023-08-231-3/+3
* Bump the crates version to 0.1.2. (#522)Laurent Mazare2023-08-201-3/+3
* Relax the requirements on CustomOp. (#486)Laurent Mazare2023-08-171-2/+2
* add c++17 flags (#452)Chengxu Yang2023-08-151-0/+1
* Rename vec-dot to vec-ops. (#449)Laurent Mazare2023-08-151-3/+3
* Add the license files. (#335)Laurent Mazare2023-08-071-1/+1
* Update the repo location. (#305)Laurent Mazare2023-08-021-1/+1
* Add some missing readme files. (#304)Laurent Mazare2023-08-021-0/+1
* Add version numbers for all the candle crates (#303)Laurent Mazare2023-08-021-2/+2
* Rename the candle crate to candle-core (#301)Laurent Mazare2023-08-021-1/+1
* Fix the flash-attention function names. (#282)Laurent Mazare2023-07-311-2/+2
* Flash attention without padding (varlen). (#281)Laurent Mazare2023-07-314-4/+283
* Softmax numerical stability. (#267)Laurent Mazare2023-07-282-1/+2
* Add some flash attn test (#253)Laurent Mazare2023-07-264-12/+123
* Use bail rather than wrapping a string where possible. (#249)Laurent Mazare2023-07-261-2/+2
* Lining up the flash attn version with the non-flash one. (#248)Laurent Mazare2023-07-261-1/+18
* Again set a few extra params in flash-attn. (#245)Laurent Mazare2023-07-2620-115/+471
* Proper flash-attn parameters. (#244)Laurent Mazare2023-07-263-8/+122
* Specific cache dir for the flash attn build artifacts. (#242)Laurent Mazare2023-07-261-10/+10
* Add flash attention (#241)Laurent Mazare2023-07-2615-0/+2655
generated by cgit v1.2.3 (git 2.39.1) at 2025-04-05 00:35:27 +0000