 Charles Xu
|
2b3efea9a4
kleidiai: fix GGML_ASSERT(*cur_backend_id != -1) failed (#15614)
|
4 ماه پیش |
 hipudding
|
c0389dba43
CANN: Disable acl_graph for prefill stage (#15933)
|
4 ماه پیش |
 Oliver Simons
|
00681dfc16
CUDA: Add `fastdiv` to `k_bin_bcast*`, giving 1-3% E2E performance (#15872)
|
4 ماه پیش |
 Jie Fu (傅杰)
|
4f658855fa
llama : support T5 models with unequal number of encoder-decoder layers (#15909)
|
4 ماه پیش |
 Sigbjørn Skjæret
|
6ab397e12b
graph : support non-contiguous Q in build_attn_mha (#15908)
|
4 ماه پیش |
 Daniel Bevenius
|
9de447d94e
ggml-cpu : fix padding in ggml_timestep_embedding (#15917)
|
4 ماه پیش |
 Georgi Gerganov
|
0f0a3c2851
metal : make the backend async (#15906)
|
4 ماه پیش |
|
Daniel Bevenius
|
33daece86b
ci : add caching for ROCm installation in release workflow (#15924)
|
4 ماه پیش |
|
Daniel Bevenius
|
e7b6d83b52
tests : filter out no-ops from coverage report (#15900)
|
4 ماه پیش |
 j-k
|
2cfef4d117
media : add transparent icon svg and png [no ci] (#15891)
|
4 ماه پیش |
 Jesse
|
09e72a037c
gitignore : Ignore vim swap files in tests (#15901)
|
4 ماه پیش |
 Chenguang Li
|
10d8b2b6b0
CANN: Add ROPE sin/cos cache for reuse (#15912)
|
4 ماه پیش |
|
Chenguang Li
|
28b5f190ef
CANN: implement LRU cache for ACL graphs (#15814)
|
4 ماه پیش |
|
Daniel Bevenius
|
86587da03b
llama : check returned fn ptrs from ggml_backend_reg_get_proc_address (#15893)
|
4 ماه پیش |
|
Daniel Bevenius
|
ff02caf9ee
ci : cache ROCm installation in windows-latest-cmake-hip (#15887)
|
4 ماه پیش |
 Ruben Ortlam
|
ae355f6f71
vulkan: throw the oom error instead of no memory type found (#15905)
|
4 ماه پیش |
 Jeff Bolz
|
4f63cd705c
vulkan: Fix OOB accesses in soft_max_back (#15861)
|
4 ماه پیش |
 Johannes Gäßler
|
17bc5a815f
HIP: use v_dot2_f32_f16 instruction for FA (#15884)
|
4 ماه پیش |
 lksj92hs
|
ed54e32558
Workaround for subgroup arithmetic failing on MoltenVK with AMD GPUs (issue 15846) (#15886)
|
4 ماه پیش |
 Aman Gupta
|
a972faebed
CUDA: Add mul_mat_id support for the mmf kernel (#15767)
|
4 ماه پیش |
|
Johannes Gäßler
|
550cf726e1
CUDA: fix GET_ROWS for large tensors (#15882)
|
4 ماه پیش |
|
Georgi Gerganov
|
c252ce67c4
contrib : add notes about merging PRs (#15881)
|
4 ماه پیش |
|
Daniel Bevenius
|
70cd37dbbe
requirements : update transformers/torch for Embedding Gemma (#15828)
|
4 ماه پیش |
 Piotr Wilkin (ilintar)
|
acc1b008cf
model-conversion : add extra debugging support for model conversion (#15877)
|
4 ماه پیش |
 Aldehir Rojas
|
7057faf64b
json : support `enum` values within `allOf` (#15830)
|
4 ماه پیش |
|
j-k
|
fe1c92cd7b
media : add llama1 icon (#15878)
|
4 ماه پیش |
|
Jeff Bolz
|
e68aa10d8f
vulkan: sort graph to allow more parallel execution (#15850)
|
4 ماه پیش |
|
Aman Gupta
|
0a16bf52e6
CUDA: generate_cu_files.py - add missing mxfp4 (#15880)
|
4 ماه پیش |
|
Jesse
|
88021565f0
chat : Deepseek V3.1 reasoning and tool calling support (OpenAI Style) (#15533)
|
4 ماه پیش |
 Xuan-Son Nguyen
|
56920f5665
server : bring back timings_per_token (#15879)
|
4 ماه پیش |
