Skip to content

Actions: ngxson/llama.cpp

Actions

CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
build.yml
1,485 workflow run results
1,485 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

CUDA: fix MMQ nwarps for AMD with warp_size==32 (#15014) CI #1844: Commit 9c35706 pushed by ngxson
1h 6m 18s master
master
1h 6m 18s
vendor : update vendored copy of google/minja (#15011) CI #1843: Commit c76b420 pushed by ngxson
1h 10m 4s master
master
1h 10m 4s
model : add hunyuan dense (#14878) CI #1842: Commit 0f5ccd6 pushed by ngxson
1h 6m 50s master
master
1h 6m 50s
opencl: add f16 for add, sub, mul, div (#14984) CI #1841: Commit 1c872f7 pushed by ngxson
1h 4m 42s master
master
1h 4m 42s
ggml : Q2k interleaving implementation - x86/x64 SIMD (#14373) CI #1840: Commit baad948 pushed by ngxson
1h 4m 57s master
master
1h 4m 57s
graph : fix equal_seq() check (#14986) CI #1839: Commit ba42794 pushed by ngxson
1h 17m 56s master
master
1h 17m 56s
docker : add cann build pipline (#14591) CI #1838: Commit 2860d47 pushed by ngxson
56m 43s master
master
56m 43s
quantize : skip tensor override when in fallback mode (#14995) CI #1837: Commit daf2dd7 pushed by ngxson
1h 7m 45s master
master
1h 7m 45s
llama : add simple option to enable CPU for MoE weights (--cpu-moe) (… CI #1836: Commit a06ed5f pushed by ngxson
1h 6m 45s master
master
1h 6m 45s
Fix params bug in diffusion example (#14993) CI #1835: Commit 7845240 pushed by ngxson
1h 9m 27s master
master
1h 9m 27s
llama : allow other bufts when overriding to CPU, add --no-repack opt… CI #1834: Commit d6818d0 pushed by ngxson
1h 14m 3s master
master
1h 14m 3s
Vulkan: Fix minor debug mode issues (#14899) CI #1833: Commit e08a988 pushed by ngxson
1h 14m 9s master
master
1h 14m 9s
mtmd : support MiniCPM-V 4.0 (#14983) CI #1832: Commit 952a47f pushed by ngxson
1h 13m 32s master
master
1h 13m 32s
server : implement universal assisted decoding (#12635) CI #1831: Commit 94933c8 pushed by ngxson
1h 33m 26s master
master
1h 33m 26s
llama : merge build_moe_ffn_from_probs function into build_moe_ffn (#… CI #1830: Commit c1dacaa pushed by ngxson
1h 22m 57s master
master
1h 22m 57s
server : add openai-style logit_bias support (#14946) CI #1829: Commit a9f77a8 pushed by ngxson
1h 26m 58s master
master
1h 26m 58s
Add LLaDA 8b Diffusion model (#14771) CI #1828: Commit 8a4a856 pushed by ngxson
1h 3m 39s master
master
1h 3m 39s
graph : reduce splits for recurrent and hybrid models (#14825) CI #1827: Commit 66625a5 pushed by ngxson
1h 9m 21s master
master
1h 9m 21s
opencl: add mul_mat_f32_f32_l4_lm and mul_mat_f16_f32_l4_lm (#14809) CI #1826: Commit 6e67254 pushed by ngxson
1h 4m 42s master
master
1h 4m 42s
quantize : fix using combined imatrix GGUFs (multiple datasets) (#14973) CI #1825: Commit e9192be pushed by ngxson
1h 3m 36s master
master
1h 3m 36s
server : add support for embd_normalize parameter (#14964) CI #1824: Commit 41e78c5 pushed by ngxson
1h 15m 5s master
master
1h 15m 5s
HIP: enable mfma mmq on gfx908 and gfx90a for select datatypes and sh… CI #1823: Commit ad4a700 pushed by ngxson
1h 0m 39s master
master
1h 0m 39s
sync : ggml CI #1822: Commit e32a4ec pushed by ngxson
59m 21s master
master
59m 21s
CUDA: skip masked KV slices for all FA kernels (#14924) CI #1821: Commit 92b8810 pushed by ngxson
1h 10m 39s master
master
1h 10m 39s
tests : update for LLAMA_SET_ROWS=1 (#14961) CI #1820: Commit 00131d6 pushed by ngxson
1h 5m 48s master
master
1h 5m 48s