 Kerfuffle
|
b4c6f46f17
Allow cmake to build ggml as a library (#1896)
|
2 lat temu |
 David Yang
|
92f20d9942
train : get raw text instead of page with html (#1905)
|
2 lat temu |
 0cc4m
|
d411968e99
opencl : support k-quants (#1836)
|
2 lat temu |
 SuperUserNameMan
|
b41b4cad6f
examples : add "simple" (#1840)
|
2 lat temu |
 Zenix
|
13fe9d2d84
cmake : add auto detection of BLAS_INCLUDE_DIRS (#1886)
|
2 lat temu |
 Johannes Gäßler
|
ac3b886953
llama : fix embd when offloading non-repeating layers (#1891)
|
2 lat temu |
 FrankHB
|
5b9ccaf104
Fixed possible macro redefinition (#1892)
|
2 lat temu |
 Borislav Stanimirov
|
9cbf50c041
build : fix and ignore MSVC warnings (#1889)
|
2 lat temu |
 Kawrakow
|
3d01122610
CUDA : faster k-quant dot kernels (#1862)
|
2 lat temu |
|
Borislav Stanimirov
|
602c748863
gitignore : add several entries specific to Visual Studio (#1888)
|
2 lat temu |
|
Johannes Gäßler
|
a09f9195be
Fixed CUDA runtime version check (#1879)
|
2 lat temu |
 Georgi Gerganov
|
bed9275617
cmake : remove whitespaces
|
2 lat temu |
 yangli2
|
c36e81da62
examples : add chat-vicuna.sh (#1854)
|
2 lat temu |
 Igor Okulist
|
3559433fec
cmake : set include path for OpenBlas (#1830)
|
2 lat temu |
 Frederik Vogel
|
69b34a0e80
swift : Package compile breaks due to ggml-metal.metal (#1831)
|
2 lat temu |
 daboe01
|
cf267d1c71
make : add train-text-from-scratch (#1850)
|
2 lat temu |
 Srinivas Billa
|
9dda13e5e1
readme : server compile flag (#1874)
|
2 lat temu |
 sandyiscool
|
37e257c48e
make : clean *.so files (#1857)
|
2 lat temu |
 Howard Su
|
64cc19b4fe
Fix the validation of main device (#1872)
|
2 lat temu |
|
Georgi Gerganov
|
4bfcc855ab
metal : parallel command buffer encoding (#1860)
|
2 lat temu |
|
Johannes Gäßler
|
6b8312e797
Better error when using both LoRA + GPU layers (#1861)
|
2 lat temu |
|
Johannes Gäßler
|
254a7a7a5f
CUDA full GPU acceleration, KV cache in VRAM (#1827)
|
2 lat temu |
 0xspringtime
|
9254920265
baby-llama : fix operator!= (#1821)
|
2 lat temu |
 xaedes
|
e32089b2c2
train : improved training-from-scratch example (#1652)
|
2 lat temu |
|
Georgi Gerganov
|
2347e45e7b
llama : do a warm-up eval at start for better timings (#1824)
|
2 lat temu |
|
Kerfuffle
|
74d4cfa343
Allow "quantizing" to f16 and f32 (#1787)
|
2 lat temu |
|
Kawrakow
|
74a6d922f1
Metal implementation for all k_quants (#1807)
|
2 lat temu |
 slaren
|
e4caa8da59
ci : run when changing only the CUDA sources (#1800)
|
2 lat temu |
|
Howard Su
|
58970a4c39
Leverage mmap for offloading tensors to GPU (#1597)
|
2 lat temu |
|
Kawrakow
|
8c0a10e64d
metal : fix failure to load model (#1817)
|
2 lat temu |
