Jeff Bolz
|
24e86cae72
vulkan: KHR_coopmat flash attention (#13506)
|
há 8 meses atrás |
Jeff Bolz
|
dc1d2adfc0
vulkan: scalar flash attention implementation (#13324)
|
há 9 meses atrás |
Jeff Bolz
|
8ae5ebcf85
vulkan: Additional type support for unary, binary, and copy (#13266)
|
há 9 meses atrás |
Georgi Gerganov
|
b34443923c
sync : ggml (#13268)
|
há 9 meses atrás |
Jeff Bolz
|
79f26e9e12
vulkan: Add bfloat16 support (#12554)
|
há 9 meses atrás |
Jeff Bolz
|
f01bd02376
vulkan: Implement split_k for coopmat2 flash attention. (#12627)
|
há 10 meses atrás |
0cc4m
|
a8a1f33567
Vulkan: Add DP4A MMQ and Q8_1 quantization shader (#12135)
|
há 10 meses atrás |
Jeff Bolz
|
eddfb43850
vulkan: Optimize mul_mat_vec p021 and nc shaders (#12505)
|
há 10 meses atrás |
stduhpf
|
4375415b4a
Vulkan: RTE rounding for cpy to quant (#12480)
|
há 10 meses atrás |
Molly Sophia
|
7dfad387e3
llama: Add support for RWKV v7 architecture (#12412)
|
há 10 meses atrás |
Rémy O
|
438a83926a
vulkan: add specific MMV kernels for IQ2 and IQ3 quants + optimizations (#11595)
|
há 11 meses atrás |
Eve
|
fbeda9002d
vulkan: matmul dequantization improvements (#12015)
|
há 11 meses atrás |
Judd
|
c132239bfb
add OP sigmoid (#12056)
|
há 11 meses atrás |
Rémy O
|
61d4f39dfe
vulkan: implement more backpropagation operators (#11914)
|
há 11 meses atrás |
Rémy O
|
2eea03d86a
vulkan: implement several ops relevant for ggml_opt (#11769)
|
há 11 meses atrás |
Jeff Bolz
|
bf42a23d0a
vulkan: support multi/vision rope, and noncontiguous rope (#11902)
|
há 11 meses atrás |
Rémy O
|
fc1b0d0936
vulkan: initial support for IQ1_S and IQ1_M quantizations (#11528)
|
há 11 meses atrás |
Rémy O
|
8a7e3bf17a
vulkan: initial support for IQ4_XS quantization (#11501)
|
há 1 ano atrás |
Rémy Oudompheng
|
66ee4f297c
vulkan: implement initial support for IQ2 and IQ3 quantizations (#11360)
|
há 1 ano atrás |
Jeff Bolz
|
1971adf55e
vulkan: sort shaders for more deterministic binary (#11315)
|
há 1 ano atrás |
Jeff Bolz
|
aea8ddd516
vulkan: fix coopmat2 validation failures (#11284)
|
há 1 ano atrás |
Jeff Bolz
|
bd38ddea01
vulkan: support copy from f32 to q4_0/q4_1/q5_0/q5_1/q8_0/iq4_nl (#11166)
|
há 1 ano atrás |
Junil Kim
|
1d8504338e
fix: ggml: fix vulkan-shaders-gen build (#10448)
|
há 1 ano atrás |
Mathieu Baudier
|
02f0430141
Disable GL_KHR_cooperative_matrix Vulkan extension if not available. (#11117)
|
há 1 ano atrás |
Peter
|
d283d02bf2
examples, ggml : fix GCC compiler warnings (#10983)
|
há 1 ano atrás |
Zhiyuan Li
|
160bc039c8
rwkv6: add wkv6 support for Vulkan backend (#10829)
|
há 1 ano atrás |
Jeff Bolz
|
b685daf386
vulkan: request round-to-even for fp16 in im2col/rope_head (#10767)
|
há 1 ano atrás |
Jeff Bolz
|
a05e2afcc2
vulkan: disable spirv-opt for coopmat shaders (#10763)
|
há 1 ano atrás |
Jeff Bolz
|
ecc93d0558
vulkan: compile a test shader in cmake to check for coopmat2 support (#10713)
|
há 1 ano atrás |
0cc4m
|
3df784b305
Vulkan: VK_KHR_cooperative_matrix support to speed up prompt processing (#10597)
|
há 1 ano atrás |