 Nicolò Scipione
|
94148ba330
sycl: allow ggml-sycl configuration and compilation using Visual Studio project/solution (#12625)
|
9 bulan lalu |
 Ronny Brendel
|
9ac4d611d0
cmake: fix ggml-shaders-gen compiler paths containing spaces (#12747)
|
9 bulan lalu |
 Daniel Bevenius
|
348888e0dc
docs : add XCFramework section to README.md [no ci] (#12746)
|
9 bulan lalu |
 Jeff Bolz
|
74d4f5b041
vulkan: Hybrid waitForFences/getFenceStatus to reduce fence latency (#12630)
|
9 bulan lalu |
|
Jeff Bolz
|
35e592eb30
vulkan: set cmake minimum and project name in vulkan-shaders (#12744)
|
9 bulan lalu |
 lhez
|
7d7b1bafa7
opencl: update doc for OpenCL (#12702)
|
9 bulan lalu |
 Gaurav Garg
|
c262beddf2
CUDA: Prefer vector flash decoding kernel for Gemma models (#12738)
|
9 bulan lalu |
 yumeyao
|
5dd5d1ab00
vocab : use string_view::find() to avoid unnecessary looking up beyond the fragment range (#12706)
|
9 bulan lalu |
|
Jeff Bolz
|
1c059995e0
vulkan: Fix missing cmake logic for dot product extension (#12721)
|
9 bulan lalu |
 Atharva Dubey
|
2004644b7a
ci : add env variable in ggml-ci and document the same in SYCL.md (#12736)
|
9 bulan lalu |
 R0CKSTAR
|
5f696e88e0
sync : minja (inclusionAI/Ling) and update tests (#12699)
|
9 bulan lalu |
 a3sh
|
193c3e03a6
fix MUSA compiler warning (#12704)
|
9 bulan lalu |
 Chenguang Li
|
65cfe136a0
CANN: Support operator SIN COS ARGMAX (#12709)
|
9 bulan lalu |
 Alan Gray
|
3f9da22c2b
Simplify and improve CUDA graphs through use of indirect copy pointers (#9017)
|
9 bulan lalu |
 hipudding
|
2a0dc97e56
CANN: Fix failed test cases (#12708)
|
9 bulan lalu |
|
lhez
|
97a20c012b
opencl: use `max_alloc_size` in backend ctx instead of querying again (#12705)
|
9 bulan lalu |
|
Jeff Bolz
|
f01bd02376
vulkan: Implement split_k for coopmat2 flash attention. (#12627)
|
9 bulan lalu |
 bandoti
|
6f3bd38640
cmake: remove caching from vulkan coopmat checks (#12719)
|
9 bulan lalu |
|
Jeff Bolz
|
be0a0f8cae
vulkan: Implement grouped query attention in the coopmat2 FA shader (#12559)
|
9 bulan lalu |
 0cc4m
|
92e3006bb6
Vulkan: Fix mmq int dot float cache size (#12722)
|
9 bulan lalu |
 Georgi Gerganov
|
833e2b7409
model : print tensor size during load (#12711)
|
9 bulan lalu |
 Diego Devesa
|
e0e912f49b
llama : add option to override model tensor buffers (#11397)
|
9 bulan lalu |
|
Georgi Gerganov
|
a10b36c91a
llama : refactor kv cache guard (#12695)
|
9 bulan lalu |
 Sigbjørn Skjæret
|
83a88bd6af
vocab : BailingMoE : change possessive quantifiers to greedy (#12677)
|
9 bulan lalu |
 Xuan-Son Nguyen
|
42eb248f46
common : remove json.hpp from common.cpp (#12697)
|
9 bulan lalu |
|
Chenguang Li
|
9bacd6b374
[CANN] get_rows and dup optimization (#12671)
|
9 bulan lalu |
|
Xuan-Son Nguyen
|
267c1399f1
common : refactor downloading system, handle mmproj with -hf option (#12694)
|
9 bulan lalu |
 Junil Kim
|
f423981ac8
opencl : fix memory allocation size (#12649)
|
9 bulan lalu |
 jklincn
|
e39e727e9a
llama : use LLM_KV_GENERAL_FILE_TYPE instead of gguf_find_key (#12672)
|
9 bulan lalu |
|
Sigbjørn Skjæret
|
5936a616e4
convert : BailingMoE : fix qkv split when head_dim is 0 (#12687)
|
9 bulan lalu |
