issues
search
ggerganov
/
ggml
Tensor library for machine learning
MIT License
10.93k
stars
1k
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
test: fix OPT_STEP_ADAMW for test-backend-ops
#974
JohannesGaessler
closed
1 hour ago
0
ggml : fix ggml_cast
#973
iboB
closed
1 hour ago
1
CUDA: remove bad assert
#972
JohannesGaessler
closed
15 hours ago
0
A specialized Winograd Conv2d op
#971
bssrdf
opened
1 day ago
7
different conv-2d results between CUDA and CPU backends
#970
bssrdf
closed
1 day ago
3
ggml : fix GGML_MAX_N_THREADS + improve formatting
#969
ggerganov
closed
5 days ago
0
bug: `GGML_MAX_N_THREADS` undefined if `GGML_MAX_NAME` is defined
#968
Green-Sky
closed
5 days ago
1
sync : llama.cpp
#967
ggerganov
closed
5 days ago
0
ggml: fix gradient allocation logic
#966
JohannesGaessler
closed
11 hours ago
15
sync : llama.cpp
#965
ggerganov
closed
1 week ago
0
vulkan: multithread pipeline creation
#963
jeffbolznv
closed
16 hours ago
6
vulkan: fix build for GGML_VULKAN_RUN_TESTS, add TFLOPS to logging
#961
jeffbolznv
closed
3 days ago
2
build: fix msvc build
#960
jeffbolznv
closed
1 week ago
0
custom blas
#959
wmx-github
opened
2 weeks ago
0
Build failed when set Android
#958
liangzelang
opened
2 weeks ago
0
Possible bug in ggml_permute
#957
bernerprzemek
closed
2 weeks ago
2
Build failed
#956
zixi01chen
opened
2 weeks ago
0
sync : llama.cpp
#955
ggerganov
closed
3 weeks ago
0
ggml_conv_2d input shape
#954
StudyingLover
opened
3 weeks ago
4
vulkan: mul_mat: fix UB with small warps
#952
smeso
closed
1 hour ago
3
vulkan: argsort: barriers must be under uniform control flow
#951
smeso
closed
4 days ago
1
ggml/examples: add backend support for numerical optimization
#949
JohannesGaessler
closed
1 week ago
21
vulkan: fix compilation with GGML_VULKAN_DEBUG=ON
#948
smeso
closed
3 weeks ago
0
vulkan: add dryrun support to sin and cos ops
#947
smeso
closed
3 weeks ago
0
vulkan: correctly report support for OP_CONT
#946
smeso
closed
3 weeks ago
2
ggml: fix ggml_graph_cpy undefined behavior
#943
JohannesGaessler
closed
4 weeks ago
0
Do we need to build the computational graph every time during prediction?
#942
EeyoreLee
closed
5 days ago
1
Error in MUL_MAT with GGML_VULKAN_CHECK_RESULTS
#941
SRHMorris
opened
1 month ago
0
Add conv_transpose_1d_gemm
#940
smeso
opened
1 month ago
8
sync : llama.cpp
#939
ggerganov
closed
1 month ago
0
Please sync Vulkan from llama.cpp
#938
SkutteOleg
closed
1 month ago
5
SYCL: do not skip CPU if explicitly selected
#937
smeso
closed
4 weeks ago
2
Fix memory leak in tests
#936
smeso
closed
1 month ago
0
Support for warp sizes of 64
#935
Haus1
opened
1 month ago
1
ggml_cont: fix issue with transposed tensors when one dimension is 1
#934
smeso
closed
1 month ago
0
contbug pr
#933
smeso
closed
1 month ago
1
tests: add gradient tests for all backends
#932
JohannesGaessler
closed
3 weeks ago
6
Different output between CPU and Metal with the im2col operator
#931
lovemefan
opened
1 month ago
4
Call log callbacks before abort
#930
thewh1teagle
opened
1 month ago
0
feat: refactor cross entropy, add CUDA, fix grad test
#929
JohannesGaessler
closed
1 month ago
0
ggml: remove bad assert
#928
JohannesGaessler
closed
1 month ago
0
`NULL` assert error in MNIST example for current version on Mac
#927
Ucag
closed
1 month ago
0
cublasGemmBatchedEx failed to launch on GPUs for some matrices
#926
bssrdf
closed
1 month ago
1
yolo : add backend support
#924
rgerganov
closed
1 month ago
0
How to run tests
#923
bhanuprakashbv
opened
1 month ago
0
ggml : fix typo in ggml-quants.c comment
#922
danbev
closed
1 month ago
0
feat: implement backward computation for more operators
#921
Ronsor
opened
1 month ago
5
sync : llama.cpp
#920
ggerganov
closed
1 month ago
0
feat: add new `sin` and `cos` operators
#919
Ronsor
closed
1 month ago
0
sync : llama.cpp
#918
ggerganov
closed
1 month ago
0
Next