Skip to content

Actions: ngxson/llama.cpp

Actions

Server

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
server.yml
690 workflow run results
690 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

vulkan: initial support for IQ4_XS quantization (#11501) Server #947: Commit 8a7e3bf pushed by ngxson
9m 36s master
master
9m 36s
vulkan: use smaller combined allocations to avoid fragmentation (#11551) Server #946: Commit 1b598b3 pushed by ngxson
7m 43s master
master
7m 43s
metal : avoid breaking build when metal API predates TARGET_OS_VISION… Server #945: Commit 902368a pushed by ngxson
7m 58s master
master
7m 58s
metal : adjust support conditions for norm operators (#11671) Server #944: Commit d774ab3 pushed by ngxson
8m 0s master
master
8m 0s
CUDA: support for mat. mul. with ne03 != ne13 (#11656) Server #943: Commit fa62da9 pushed by ngxson
17m 26s master
master
17m 26s
llava: add quantization for the visual projector LLAVA, Qwen2VL (#11644) Server #942: Commit 1ec2080 pushed by ngxson
7m 49s master
master
7m 49s
sync: minja (#11641) Server #941: Commit 9f4cc8f pushed by ngxson
8m 35s master
master
8m 35s
CUDA: non-contiguous (RMS) norm support (#11659) Server #940: Commit fd08255 pushed by ngxson
8m 54s master
master
8m 54s
HIP: force max threads per block to be 1024 (#11621) Server #939: Commit 3ec9fd4 pushed by ngxson
9m 7s master
master
9m 7s
server : add try..catch to places not covered by set_exception_handle… Server #938: Commit 3962fc1 pushed by ngxson
7m 51s master
master
7m 51s
arg : list RPC devices first when using --list-devices (#11655) Server #937: Commit 1bef571 pushed by ngxson
8m 3s master
master
8m 3s
tool-call: command r7b fix for normal responses (#11608) Server #936: Commit db288b6 pushed by ngxson
7m 52s master
master
7m 52s
swift : fix llama-vocab api usage (#11645) Server #935: Commit f117d84 pushed by ngxson
19m 0s master
master
19m 0s
metal : use residency set for other platforms (#11648) Server #934: Commit 534c46b pushed by ngxson
8m 38s master
master
8m 38s
sync : ggml Server #933: Commit 7c9e0ca pushed by ngxson
8m 43s master
master
8m 43s
tool-call: allow --chat-template chatml w/ --jinja, default to … Server #932: Commit cde3833 pushed by ngxson
11m 50s master
master
11m 50s
server : (webui) revert hacky solution from #11626 (#11634) Server #931: Commit b345178 pushed by ngxson
7m 51s master
master
7m 51s
server : (webui) allow typing and submitting during llm response (#11… Server #930: Commit 1d1e6a9 pushed by ngxson
8m 42s master
master
8m 42s
server : remove CPPHTTPLIB_NO_EXCEPTIONS define (#11622) Server #929: Commit 5598f47 pushed by ngxson
8m 10s master
master
8m 10s
CUDA: fix Volta FlashAttention logic (#11615) Server #928: Commit 21c84b5 pushed by ngxson
12m 35s master
master
12m 35s
server : (webui) Fix Shift+Enter handling (#11609) Server #927: Commit d92cb67 pushed by ngxson
8m 0s master
master
8m 0s
HIP: fix flash_attn_stream_k_fixup warning (#11604) Server #926: Commit 6eecde3 pushed by ngxson
7m 43s master
master
7m 43s
CUDA/HIP: add support for selectable warp size to mmv (#11519) Server #925: Commit 396856b pushed by ngxson
7m 18s master
master
7m 18s
HIP: add GGML_CUDA_CC_IS_* for amd familys as increasing cc archtectu… Server #924: Commit 4d0598e pushed by ngxson
7m 35s master
master
7m 35s
nit: more informative crash when grammar sampler fails (#11593) Server #923: Commit 90f9b88 pushed by ngxson
9m 52s master
master
9m 52s
ProTip! You can narrow down the results and go further in time using created:<2025-02-02 or the other filters available.