 Daniel Bevenius
|
ca0ef2dddb
llama : clarify comment about pp and tg graphs [no ci] (#14895)
|
5 месяцев назад |
 Erik Scholz
|
89d1029559
vulkan : add fp16 support for the conv_2d kernel (#14872)
|
5 месяцев назад |
 Jeff Bolz
|
f1a4e72de5
vulkan: skip empty set_rows to avoid invalid API usage (#14860)
|
5 месяцев назад |
 Gabriel Larson
|
4762ad7316
model : make rope_yarn_log_mul optional for deepseek2 (#14896)
|
5 месяцев назад |
 Shunta Saito
|
1dc9614e06
llama : fix kq_scale for the attention layers of PLaMo2 (#14892)
|
5 месяцев назад |
 Aman Gupta
|
446595b9b3
Docs: add instructions for adding backends (#14889)
|
5 месяцев назад |
 deepsek
|
66906cd82a
HIP: Enable Matrix cores for MMQ Kernels, Enable stream-K for CDNA 3 (#14624)
|
5 месяцев назад |
 hipudding
|
11dd5a44eb
CANN: Implement GLU ops (#14884)
|
5 месяцев назад |
 R0CKSTAR
|
9b8f3c6c77
musa: fix build warnings (unused variable) (#14869)
|
5 месяцев назад |
 Aaron Teo
|
c7f3169cd5
ggml-cpu : disable GGML_NNPA by default due to instability (#14880)
|
5 месяцев назад |
 Gabe Goodhart
|
793c0d7f46
metal: SSM_SCAN performance (#14743)
|
5 месяцев назад |
 lhez
|
ce111d39d6
opencl: add fused `rms_norm_mul` (#14841)
|
5 месяцев назад |
 wooksong
|
e7fecba934
docs : update HOWTO‑add‑model.md for ModelBase and new model classes (#14874)
|
5 месяцев назад |
 Oliver Simons
|
e2b7621e7c
ggml : remove invalid portPos specifiers from dot files (#14838)
|
5 месяцев назад |
 Georgi Gerganov
|
c1dbea752a
context : restore preemptive sched reset when LLAMA_SET_ROWS=0 (#14870)
|
5 месяцев назад |
 kiwi
|
749e0d27f0
mtmd : fix 32-bit narrowing issue in export-lora and mtmd clip (#14503)
|
5 месяцев назад |
 Chris Rohlf
|
64bf1c3744
rpc : check for null buffers in get/set/copy tensor endpoints (#14868)
|
5 месяцев назад |
 Diego Devesa
|
c12bbde372
sched : fix multiple evaluations of the same graph with pipeline parallelism (#14855)
|
5 месяцев назад |
|
R0CKSTAR
|
3f4fc97f1d
musa: upgrade musa sdk to rc4.2.0 (#14498)
|
5 месяцев назад |
|
Georgi Gerganov
|
2df255da3c
sync : ggml
|
5 месяцев назад |
 Kai Pastor
|
60f816a79d
cmake : fix usage issues (ggml/1257)
|
5 месяцев назад |
|
Daniel Bevenius
|
5592f278b6
ggml-cpu : remove stdlib include from repack.cpp (ggml/1276)
|
6 месяцев назад |
|
Georgi Gerganov
|
e4868d16d2
context : perform output reorder lazily upon access after sync (#14853)
|
5 месяцев назад |
 Xuan-Son Nguyen
|
820de57d4f
chat : fix kimi-k2 chat template (#14852)
|
5 месяцев назад |
 Alberto Cabrera Pérez
|
cb4a63aad6
sycl: fixed semantics of block offset calculation (#14814)
|
5 месяцев назад |
 yummy
|
86f5623d90
llama : fix MiniCPM inference after Granite Four changes (#14850)
|
5 месяцев назад |
 Pouya
|
39cffdf188
docs: add libcurl-dev install hint for Linux distros (#14801)
|
5 месяцев назад |
|
Georgi Gerganov
|
065908cb09
metal : fix fusion across different encoders (#14849)
|
5 месяцев назад |
 Donghyeon Jeong
|
4ec6291a24
sycl: fix undefined variable in work group size check (#14843)
|
5 месяцев назад |
 jacekpoplawski
|
a12363bbf0
convert : text-only support for GLM-4.1V-9B-Thinking (#14823)
|
5 месяцев назад |
