Skip to content

Actions: ngxson/llama.cpp

Actions

CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
build.yml
786 workflow run results
786 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

CUDA: Prefer vector flash decoding kernel for Gemma models (#12738) CI #1140: Commit c262bed pushed by ngxson
48m 56s master
master
48m 56s
vocab : use string_view::find() to avoid unnecessary looking up beyon… CI #1139: Commit 5dd5d1a pushed by ngxson
45m 11s master
master
45m 11s
vulkan: Fix missing cmake logic for dot product extension (#12721) CI #1138: Commit 1c05999 pushed by ngxson
45m 17s master
master
45m 17s
sync : minja (inclusionAI/Ling) and update tests (#12699) CI #1137: Commit 5f696e8 pushed by ngxson
45m 21s master
master
45m 21s
fix MUSA compiler warning (#12704) CI #1136: Commit 193c3e0 pushed by ngxson
51m 1s master
master
51m 1s
CANN: Support operator SIN COS ARGMAX (#12709) CI #1135: Commit 65cfe13 pushed by ngxson
51m 53s master
master
51m 53s
Simplify and improve CUDA graphs through use of indirect copy pointer… CI #1134: Commit 3f9da22 pushed by ngxson
54m 18s master
master
54m 18s
CANN: Fix failed test cases (#12708) CI #1133: Commit 2a0dc97 pushed by ngxson
48m 32s master
master
48m 32s
opencl: use max_alloc_size in backend ctx instead of querying again… CI #1132: Commit 97a20c0 pushed by ngxson
47m 58s master
master
47m 58s
vulkan: Implement split_k for coopmat2 flash attention. (#12627) CI #1131: Commit f01bd02 pushed by ngxson
47m 55s master
master
47m 55s
cmake: remove caching from vulkan coopmat checks (#12719) CI #1130: Commit 6f3bd38 pushed by ngxson
50m 36s master
master
50m 36s
vulkan: Implement grouped query attention in the coopmat2 FA shader (… CI #1129: Commit be0a0f8 pushed by ngxson
49m 18s master
master
49m 18s
Vulkan: Fix mmq int dot float cache size (#12722) CI #1128: Commit 92e3006 pushed by ngxson
46m 49s master
master
46m 49s
model : print tensor size during load (#12711) CI #1127: Commit 833e2b7 pushed by ngxson
50m 19s master
master
50m 19s
llama : add option to override model tensor buffers (#11397) CI #1126: Commit e0e912f pushed by ngxson
45m 48s master
master
45m 48s
vocab : BailingMoE : change possessive quantifiers to greedy (#12677) CI #1125: Commit 83a88bd pushed by ngxson
44m 14s master
master
44m 14s
common : remove json.hpp from common.cpp (#12697) CI #1124: Commit 42eb248 pushed by ngxson
49m 19s master
master
49m 19s
[CANN] get_rows and dup optimization (#12671) CI #1123: Commit 9bacd6b pushed by ngxson
50m 7s master
master
50m 7s
opencl : fix memory allocation size (#12649) CI #1122: Commit f423981 pushed by ngxson
43m 59s master
master
43m 59s
llama : use LLM_KV_GENERAL_FILE_TYPE instead of gguf_find_key (#12672) CI #1121: Commit e39e727 pushed by ngxson
43m 50s master
master
43m 50s
metal : use F32 prec in FA kernels (#12688) CI #1120: Commit 3fd072a pushed by ngxson
45m 4s master
master
45m 4s
Fix clang warning in gguf_check_reserved_keys (#12686) CI #1119: Commit a6f32f0 pushed by ngxson
55m 41s master
master
55m 41s
vulkan: fix build when glslc doesn't support coopmat (#12683) CI #1118: Commit 2bb3597 pushed by ngxson
46m 52s master
master
46m 52s
SYCL: Rename oneMKL to oneMath (#12192) CI #1117: Commit 8293970 pushed by ngxson
56m 1s master
master
56m 1s
SYCL: switch to SYCL namespace (#12674) CI #1116: Commit 8bbf260 pushed by ngxson
45m 27s master
master
45m 27s