 slaren
|
344f9126cc
ggml : tag ggml_tensor::backend as deprecated (#7290)
|
hai 1 ano |
 John Balis
|
48aa8fd1f2
ggml : add `ggml_upscale_ext` (ggml/814)
|
hai 1 ano |
 Georgi Gerganov
|
e8a7fd4fb0
metal : support FA without mask + add asserts (#7278)
|
hai 1 ano |
|
Georgi Gerganov
|
c3c88f296a
ggml : try fix ppc64 (whisper/0)
|
hai 1 ano |
|
Georgi Gerganov
|
325756d28d
ggml : resolve merge (ggml/0)
|
hai 1 ano |
 Justina Cho
|
f5ef34e428
feat: implemented sigmoid function (ggml/806)
|
hai 1 ano |
|
Georgi Gerganov
|
9cb317f77e
ggml : full ALiBi support (#7192)
|
hai 1 ano |
 Justine Tunney
|
3855416027
ggml : introduce bfloat16 support (#6412)
|
hai 1 ano |
 Xuan Son Nguyen
|
842500144e
gguf-split: add --no-tensor-first-split (#7072)
|
hai 1 ano |
|
Georgi Gerganov
|
9c67c2773d
ggml : add Flash Attention (#5021)
|
hai 1 ano |
|
Xuan Son Nguyen
|
7bb36ccf91
gguf : enforce that tensor names are unique (#6905)
|
hai 1 ano |
|
slaren
|
e2764cd7ca
gguf : fix mismatch between alloc and free functions (#6929)
|
hai 1 ano |
|
Georgi Gerganov
|
83b72cb086
Merge pull request from GHSA-p5mv-gjc5-mwqv
|
hai 1 ano |
|
Georgi Gerganov
|
51543729ff
ggml : fix redefinition of vaddvq_f32 for 32-bit ARM (#6906)
|
hai 1 ano |
|
Justine Tunney
|
192090bae4
llamafile : improve sgemm.cpp (#6796)
|
hai 1 ano |
|
slaren
|
0d56246f4b
ggml : group all experts in a single ggml_mul_mat_id (#6505)
|
hai 1 ano |
|
Georgi Gerganov
|
666867b799
ggml : fix llamafile sgemm wdata offsets (#6710)
|
hai 1 ano |
|
Justine Tunney
|
8cc91dc63c
ggml : add llamafile sgemm (#6414)
|
hai 1 ano |
|
slaren
|
fbbc030ba9
metal : unify mul_mv_id kernels (#6556)
|
hai 1 ano |
 jiez
|
91c736015b
llama : add gguf_remove_key + remove split meta during quantize (#6591)
|
hai 1 ano |
 Carolinabanana
|
5dc9dd7152
llama : add Command R Plus support (#6491)
|
hai 1 ano |
|
slaren
|
08a0c02060
ggml : mul_mat_id use the same tensor for all the experts (#6387)
|
hai 1 ano |
 0cc4m
|
ba0c7c70ab
Vulkan k-quant mmq and ggml-backend offload functionality (#6155)
|
hai 1 ano |
|
slaren
|
e5b89a441a
ggml : fix bounds checking of zero size views (#6347)
|
hai 1 ano |
 compilade
|
557410b8f0
llama : greatly reduce output buffer memory usage (#6122)
|
hai 1 ano |
 Kawrakow
|
55c1b2a3bb
IQ1_M: 1.75 bpw quantization (#6302)
|
hai 1 ano |
|
slaren
|
280345968d
cuda : rename build flag to LLAMA_CUDA (#6299)
|
hai 1 ano |
 Rick G
|
a32b77c4b2
Fix heap corruption from wmode out-of-bound writes on windows (#6272)
|
hai 1 ano |
 Meng, Hengyu
|
ddf6568510
[SYCL] offload op (#6217)
|
hai 1 ano |
 Jared Van Bortel
|
94d1b3b411
use _wfopen instead of fopen on Windows (#6248)
|
hai 1 ano |
