 Markus Tavenrath
|
daa9623ab0
Overlap cmdbuffer creation and cmdbuffer execution in Vulkan backend by submitting smaller cmdbuffers early. (#9118)
|
1 anno fa |
 Salvatore Mesoraca
|
406c1a32a1
vulkan: add dryrun support to sin and cos ops (ggml/947)
|
1 anno fa |
|
Salvatore Mesoraca
|
9cb9260861
vulkan: correctly report support for OP_CONT (ggml/946)
|
1 anno fa |
 Changyeon Kim
|
409dc4f8bb
ggml : fix build break for the vulkan-debug (#9265)
|
1 anno fa |
 Georgi Gerganov
|
231cff5f6f
sync : ggml
|
1 anno fa |
|
Changyeon Kim
|
2f3c1466ff
llava: Add ACC OP for GPU acceleration to the Vulkan backend in the LLAVA CLIP model. (#8984)
|
1 anno fa |
 0cc4m
|
5fd89a70ea
Vulkan Optimizations and Fixes (#8959)
|
1 anno fa |
 Daniel Bevenius
|
06943a69f6
ggml : move rope type enum to ggml.h (#8949)
|
1 anno fa |
|
Markus Tavenrath
|
7c5bfd57f8
Optimize Vulkan backend for better CPU performance and less GPU synchronization overhead. (#8943)
|
1 anno fa |
 Matt Stephenson
|
70c0ea3560
whisper : use vulkan as gpu backend when available (whisper/2302)
|
1 anno fa |
|
0cc4m
|
a3738b2fa7
vulkan : implement Stable Diffusion operators (ggml/904)
|
1 anno fa |
 Tony Wasserka
|
203b7f1531
vulkan : initialize vk_buffer_struct members to VK_NULL_HANDLE (ggml/893)
|
1 anno fa |
 slaren
|
2b1f616b20
ggml : reduce hash table reset cost (#8698)
|
1 anno fa |
|
0cc4m
|
751fcfc6c3
Vulkan IQ4_NL Support (#8613)
|
1 anno fa |
|
0cc4m
|
bda62d7999
Vulkan MMQ Fix (#8479)
|
1 anno fa |
|
Georgi Gerganov
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 anno fa |
