 Ruben Ortlam
|
dff7551bfd
vulkan: fix mmv subgroup16 selection (#15775)
|
há 4 meses atrás |
 Jeff Bolz
|
0fce7a1248
vulkan: don't use std::string in load_shaders, to improve compile time (#15724)
|
há 4 meses atrás |
 Daniel Bevenius
|
8227695d7a
vulkan : update ggml_vk_instance_validation_ext_available (#15666)
|
há 4 meses atrás |
 Shin-myoung-serp
|
0014fb4add
ggml vulkan: add hardsigmoid and hardswish operations (#15762)
|
há 4 meses atrás |
 Oliver Simons
|
661ae31c9c
CUDA: Optimize `rms_norm_f32` kernel and its fused variants, giving 1-6% perf E2E (#15715)
|
há 4 meses atrás |
|
Daniel Bevenius
|
407c23786d
model-conversion : fix pyright errors (#15770)
|
há 4 meses atrás |
 Georgi Gerganov
|
cdedb70a99
sampling : optimize dist sampler (#15704)
|
há 4 meses atrás |
|
Daniel Bevenius
|
2c8dac72eb
llama : fix incorrect model type for Gemma 270M (#15764)
|
há 4 meses atrás |
|
Daniel Bevenius
|
40a751ea9a
model-conversion : remove hardcoded /bin/bash shebangs [no ci] (#15765)
|
há 4 meses atrás |
 hipudding
|
5eae934883
CANN: Add RoPE contiguous check for 310I DUP device (#15735)
|
há 4 meses atrás |
 xctan
|
05c0380f2a
ggml-cpu : optimize RVV kernels (#15720)
|
há 4 meses atrás |
|
Daniel Bevenius
|
8c3fdf44ec
model-conversion : add missing curl script [no ci] (#15761)
|
há 4 meses atrás |
|
hipudding
|
f6da8cb86a
CANN: Mask unsupported TRANSPOSE_1D operator (#15733)
|
há 4 meses atrás |
 Chenguang Li
|
8a2234ea0c
CANN: Fix type float_t to float (#15736)
|
há 4 meses atrás |
 SnA1lGo
|
3de008208b
fix: resolve unsigned int initialization warning for n_dims/size in gguf.cpp (#15754)
|
há 4 meses atrás |
|
Oliver Simons
|
69db8a52e6
chore: Update `.clang-format` to use `BinPackArguments=true` (#15744)
|
há 4 meses atrás |
 Johannes Gäßler
|
c466abe158
llama: -fa 1/0/-1 aliases for -fa on/off/auto (#15746)
|
há 4 meses atrás |
|
Ruben Ortlam
|
0a2a3841e8
vulkan: fix shaders gen when no integer dot is available (#15740)
|
há 4 meses atrás |
|
hipudding
|
9961d244f2
CANN: Resolve soft_max precision issue (#15730)
|
há 4 meses atrás |
|
Jeff Bolz
|
25f1045f07
vulkan: Fix macro parameter order for f32 matmul shaders (#15716)
|
há 4 meses atrás |
 rmatif
|
97669e4073
opencl: add attn sinks support for FA kernels (#15706)
|
há 4 meses atrás |
|
Chenguang Li
|
2f853687b3
CANN: Support eager execution mode under ACL graph compilation (#15712)
|
há 4 meses atrás |
|
hipudding
|
ef2af57ddf
CANN: Support ext_factor in rope (#15710)
|
há 4 meses atrás |
|
Johannes Gäßler
|
5d804a4938
ggml-backend: raise GGML_MAX_SPLIT_INPUTS (#15722)
|
há 4 meses atrás |
 Gilad S.
|
d4d8dbe383
vulkan: use memory budget extension to read memory usage (#15545)
|
há 4 meses atrás |
|
Jeff Bolz
|
35a42edac8
vulkan: add missing clamps in new mul_mat_id paths (#15702)
|
há 4 meses atrás |
|
Ruben Ortlam
|
fec7911f8f
vulkan: disable large mmv subgroups on older Nvidia GPUs (#15717)
|
há 4 meses atrás |
 s-goto-11
|
078ce23ea7
ggml: SVE support for exponential functions (#15145)
|
há 4 meses atrás |
 Prashant Vithule
|
a0c2b207c5
ggml: aarch64: Implement SVE F16 kernels for vector functions (#15115)
|
há 4 meses atrás |
 Jie Fu (傅杰)
|
4b20d8b7e3
convert : remove redundant code (#15708)
|
há 4 meses atrás |
