Skip to content

Actions: ngxson/llama.cpp

Actions

CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
build.yml
1,452 workflow run results
1,452 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

cuda: remove linking to cublasLt (#14790) CI #1765: Commit 48b86c4 pushed by ngxson
1h 0m 3s master
master
1h 0m 3s
opencl: add conv2d kernel (#14403) CI #1764: Commit 6c9ee3b pushed by ngxson
1h 2m 1s master
master
1h 2m 1s
sycl: Fix im2col (#14797) CI #1763: Commit cd465d8 pushed by ngxson
1h 2m 53s master
master
1h 2m 53s
kleidiai: add support for get_rows (#14676) CI #1762: Commit 9220426 pushed by ngxson
59m 36s master
master
59m 36s
vulkan/cuda: Fix im2col when KW!=KH (#14789) CI #1761: Commit c2e058f pushed by ngxson
1h 6m 52s master
master
1h 6m 52s
llama : fix --reverse-prompt crashing issue (#14794) CI #1760: Commit c82d48e pushed by ngxson
1h 1m 52s master
master
1h 1m 52s
server : add parse_special option to /tokenize endpoint (#14783) CI #1759: Commit b4efd77 pushed by ngxson
57m 49s master
master
57m 49s
ggml: adds CONV_2D op and direct GEMM Vulkan implementation (#14316) CI #1758: Commit a979ca2 pushed by ngxson
57m 36s master
master
57m 36s
imatrix : use GGUF to store importance matrices (#9400) CI #1757: Commit 9008328 pushed by ngxson
1h 8m 45s master
master
1h 8m 45s
vulkan: Add logging for bf16 features to ggml_vk_print_gpu_info (#132… CI #1756: Commit d4b91ea pushed by ngxson
1h 0m 12s master
master
1h 0m 12s
Vulkan: Fix fprintf format-security warning (#14770) CI #1755: Commit 83f5872 pushed by ngxson
57m 3s master
master
57m 3s
metal : fuse add, mul + add tests (#14596) CI #1754: Commit bf9087f pushed by ngxson
1h 5m 58s master
master
1h 5m 58s
graph : fix graph reuse reset of params (#14760) CI #1753: Commit 9fb1042 pushed by ngxson
55m 27s master
master
55m 27s
parallel : add option for different RNG seeds (#14757) CI #1752: Commit 2adf8d8 pushed by ngxson
53m 59s master
master
53m 59s
cuda : Fix Gemma3n not executed as CUDA_GRAPH on NVGPUs (#14741) CI #1751: Commit 021cc28 pushed by ngxson
1h 11m 21s master
master
1h 11m 21s
graph : avoid huge warm-up graphs for MoE models (#14753) CI #1750: Commit d498af3 pushed by ngxson
1h 15m 42s master
master
1h 15m 42s
model : fix build after merge conflict (#14754) CI #1749: Commit eacdeb5 pushed by ngxson
59m 57s master
master
59m 57s
model : add EXAONE 4.0 support (#14630) CI #1748: Commit e0cb5c5 pushed by ngxson
28m 15s master
master
28m 15s
CUDA: set_rows + cpy.cu refactor (#14712) CI #1747: Commit f9a31ee pushed by ngxson
1h 0m 30s master
master
1h 0m 30s
graph : refactor context to not pass gf explicitly (#14629) CI #1746: Commit 8f974bc pushed by ngxson
1h 7m 19s master
master
1h 7m 19s
graph : Pass the graph placeholder message in debug mode (#14748) CI #1745: Commit 09651d0 pushed by ngxson
1h 7m 31s master
master
1h 7m 31s
use max work group size for device to replace the magic number (#14732) CI #1744: Commit 349ea79 pushed by ngxson
1h 0m 37s master
master
1h 0m 37s
model: add Ernie 4.5 MoE support (#14658) CI #1743: Commit cb887f1 pushed by ngxson
1h 8m 58s master
master
1h 8m 58s
kv-cache : fix k-shift for multiple streams (#14742) CI #1742: Commit d6fb3f6 pushed by ngxson
1h 1m 16s master
master
1h 1m 16s
llama : reuse compute graphs (#14482) CI #1741: Commit 01612b7 pushed by ngxson
1h 5m 38s master
master
1h 5m 38s