Max Krasnyansky
|
f0c7b5edf8
threads: improve ggml_barrier scaling with large number of threads (#9598)
|
1 жил өмнө |
Johannes Gäßler
|
424c5d00a9
ggml/examples: add backend support for numerical optimization (ggml/949)
|
1 жил өмнө |
Georgi Gerganov
|
a6809c6a2e
examples : add null threadpool args where needed (ggml/0)
|
1 жил өмнө |
slaren
|
64c6af3195
ggml : fix n_threads_cur initialization with one thread (#9538)
|
1 жил өмнө |
Max Krasnyansky
|
0226613853
threadpool : skip polling for unused threads (#9461)
|
1 жил өмнө |
slaren
|
23e0d70bac
ggml : move common CPU backend impl to new header (#9509)
|
1 жил өмнө |
Yuri Khrustalev
|
822b6322de
ggml : ggml_type_name return "NONE" for invalid values (#9458)
|
1 жил өмнө |
Ahmad Tameem
|
2b00fa7997
riscv : modify Makefile and add a RISCV_VECT to print log info (#9442)
|
1 жил өмнө |
Georgi Gerganov
|
d6a04f872d
ggml : hide ggml_object, ggml_cgraph, ggml_hash_set (#9408)
|
1 жил өмнө |
Radoslav Gerganov
|
293bebe077
rpc : fix segfault with nkvo (#9389)
|
1 жил өмнө |
Johannes Gäßler
|
202084d31d
tests: add gradient tests for all backends (ggml/932)
|
1 жил өмнө |
Johannes Gäßler
|
dbbebcab33
ggml: fix ggml_graph_cpy undefined behavior (ggml/943)
|
1 жил өмнө |
Salvatore Mesoraca
|
efe6a83e30
ggml : fix cont with transposed tensors when one dimension is 1 (ggml/934)
|
1 жил өмнө |
slaren
|
e32d0816ed
ggml : always check bounds on get_rows operations (#9354)
|
1 жил өмнө |
Xuan Son Nguyen
|
947538acb8
ggml : fix missing `cpu_set_t` on emscripten (#9336)
|
1 жил өмнө |
compilade
|
9bc6db28d0
ggml-quants : ternary packing for TriLMs and BitNet b1.58 (#8151)
|
1 жил өмнө |
yuri@FreeBSD
|
f771d064a9
ggml : add pthread includes on FreeBSD (#9258)
|
1 жил өмнө |
Molly Sophia
|
8f1d81a0b6
llama : support RWKV v6 models (#8980)
|
1 жил өмнө |
Faisal Zaghloul
|
42c76d1358
Threadpool: take 2 (#8672)
|
1 жил өмнө |
Georgi Gerganov
|
231cff5f6f
sync : ggml
|
1 жил өмнө |
Georgi Gerganov
|
fc18425b6a
ggml : add SSM Metal kernels (#8546)
|
1 жил өмнө |
Johannes Gäßler
|
e11bd856d5
CPU/CUDA: Gemma 2 FlashAttention support (#8542)
|
1 жил өмнө |
compilade
|
a1631e53f6
llama : simplify Mamba with advanced batch splits (#8526)
|
1 жил өмнө |
Daniel Bevenius
|
06943a69f6
ggml : move rope type enum to ggml.h (#8949)
|
1 жил өмнө |
DavidKorczynski
|
df5478fbea
ggml: fix div-by-zero (#9003)
|
1 жил өмнө |
Georgi Gerganov
|
b72942fac9
Merge commit from fork
|
1 жил өмнө |
Borislav Stanimirov
|
f93d49ab1e
ggml : ignore more msvc warnings (ggml/906)
|
1 жил өмнө |
Molly Sophia
|
2d5dd7bb3f
ggml : add epsilon as a parameter for group_norm (#8818)
|
1 жил өмнө |
Justine Tunney
|
b9dfc25ca3
ggml : fix overflows in elu function (#8866)
|
1 жил өмнө |
jdomke
|
76614f352e
ggml : reading the runtime sve config of the cpu (#8709)
|
1 жил өмнө |