Superminaren/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-09-07 20:47:30 +02:00
Code Issues Packages Projects Releases Wiki Activity
980 Commits 780 Branches 7,352 Tags
master-703ef9c
Commit Graph
8 Commits
Author SHA1 Message Date
Adam Treat 703ef9c125 Set the singleton to nullptr here. 2023-09-14 16:38:28 -04:00
Adam Treat 7ff671e149 Only use vulkan with known quant that work. 2023-09-14 09:58:28 -04:00
Aaron Miller e5ab32aab8 vulkan: disambiguate gpus with the same name 2023-09-13 12:27:40 -07:00
Adam Treat 2f7732b667 Throw an exception when allocation fails for vulkan. 2023-09-13 10:33:44 -04:00
Adam Treat 0412ec287c Completely revamp how we do object management with the vulkan backend and 
stop using so many static objects so we can tear down and bring up vulkan
on new devices in the same runtime.
 2023-09-12 14:24:49 -04:00
Adam Treat 5b2d8236a7 Switch to a dynamic dispatch table instead of linking hard against libvulkan. 2023-09-12 14:24:49 -04:00
Adam Treat ced231980e Remove warning which fails on windows. 2023-08-30 14:33:31 -04:00
niansa 4cdaa3c9cb Nomic vulkan backend licensed under the Software for Open Models License (SOM), version 1.0. 2023-08-30 10:11:01 -04:00