Spaces:
Running
Running
Commit History
talk-llama : sync llama.cpp
a40d0a7
sync : ggml
96e8b15
ggml-backend : fix async copy from CPU (llama/8897)
050174c
slaren
commited on
Updated SYCL device filtering (llama/8901)
64976cd
Ouadie EL FAROUKI
commited on
CUDA/HIP: fix tests/test-backend-ops (llama/8896)
f14c1ad
CUDA: fix padding logic for FP16/FP32 (llama/8884)
643bcdb
ggml : add epsilon as a parameter for group_norm (llama/8818)
d003891
ggml : fix overflows in elu function (llama/8866)
a12468a
Justine Tunney
commited on
ggml : reading the runtime sve config of the cpu (llama/8709)
c26339f
jdomke
domke
commited on
Fix conversion of unnormalized BF16->BF16 weights (llama/7843)
8b10f59
Fixing wrong VDR iq4nl value (llama/8812)
30eb7bc
Ouadie EL FAROUKI
commited on
ggml-cuda: Adding support for unified memory (llama/8035)
686bb18
Build: Only include execinfo.h on linux systems that support it (llama/8783)
0019ddb
Alex O'Connell
commited on
cuda : fix dmmv cols requirement to 2*GGML_CUDA_DMMV_X (llama/8800)
73e80d1
slaren
commited on
added android implementation of ggml_print_backtrace_symbols (llama/8751)
314d58a
cann: update cmake (llama/8765)
345a58d
Add `TIMESTEP_EMBEDDING` OP (llama/8707)
52eea23
ggml: bugfix: fix the inactive elements is agnostic for risc-v vector (llama/8748)
6989631
cuda : organize vendor-specific headers into vendors directory (llama/8746)
ec2f307
R0CKSTAR
commited on
add conv support (llama/8688)
f0d6f5c
feat: Support Moore Threads GPU (llama/8383)
a35db11
ggml : ignore more msvc warnings (ggml/906)
1b11fde
metal : fix struct name (ggml/912)
14cf8db
metal : add abort callback (ggml/905)
b822172
vulkan : implement Stable Diffusion operators (ggml/904)
124c156
ggml : move c parameter comment to ggml_rope_ext (ggml/901)
6d34596
ggml : resolve sync conflicst (ggml/0)
82658f5
common : handle new quant types (ggml/0)
53bb541
ggml : add ggml-aarch64 (ggml/0)
0062819
Dibakar Gope
commited on
ggml : reduce hash table reset cost (llama/8698)
9808fbf
slaren
commited on
ggml: handle ggml_init failure to fix NULL pointer deref (llama/8692)
dc51517
DavidKorczynski
commited on
fix multi-gpu issue on sycl (llama/8554)
94a6436
ggml : add and use ggml_cpu_has_llamafile() (llama/8664)
efcca56
Re-add erroneously removed -fsycl from GGML_EXTRA_LIBS (llama/8667)
6e12dfd
Joe Todd
commited on
sycl : Add support for non-release DPC++ & oneMKL (llama/8644)
2a5814c
Joe Todd
commited on
Vulkan IQ4_NL Support (llama/8613)
899145d
Allow all RDNA2 archs to use sdot4 intrinsic (llama/8629)
1d65fea
Jeroen Mostert
commited on
fix scratch size of softmax (llama/8642)
6519fd2
ggml: fix compile error for RISC-V (llama/8623)
4eec44b
Mark Zhuang
commited on
CUDA: MMQ code deduplication + iquant support (llama/8495)
6d14124
gguf : handle null name during init (llama/8587)
2f95156
ggml : fix quant dot product with odd number of blocks (llama/8549)
0083f96
ggml : add friendlier error message to fopen errors (llama/8575)
ab5b4e0
CUDA: fix partial offloading for ne0 % 256 != 0 (llama/8572)
afc137c
cmake : install all ggml public headers (llama/8480)
73a16f3
65a
65a
commited on
