Skip to content

Actions: ngxson/llama.cpp

Actions

CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
build.yml
233 workflow run results
233 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

sync : ggml CI #439: Commit 815fe72 pushed by ngxson
1h 11m 28s master
master
1h 11m 28s
build: fix build error in Windows env with OneAPI setup (#10107) CI #438: Commit e597e50 pushed by ngxson
1h 10m 58s master
master
1h 10m 58s
llama : improve output buffer type selection (#10098) CI #437: Commit 85679d3 pushed by ngxson
1h 7m 6s master
master
1h 7m 6s
llama : fix buffer checks for mamba and rwk (#10111) CI #436: Commit c02e5ab pushed by ngxson
1h 2m 8s master
master
1h 2m 8s
loader: refactor tensor weights storage (#9935) CI #435: Commit ab3d71f pushed by ngxson
1h 9m 10s master
master
1h 9m 10s
server : include scheme when printing URL (#10106) CI #434: Commit 0a683e8 pushed by ngxson
1h 13m 49s master
master
1h 13m 49s
ggml : check tensor name lengths in gguf files (#10100) CI #433: Commit dea5e86 pushed by ngxson
1h 16m 39s master
master
1h 16m 39s
kompute: add mul_mat_q4_k shader (#10097) CI #432: Commit 1329c0a pushed by ngxson
1h 8m 26s master
master
1h 8m 26s
kompute: add backend registry / device interfaces (#10045) CI #431: Commit 61408e7 pushed by ngxson
1h 14m 56s master
master
1h 14m 56s
ggml : fix memory leaks when loading invalid gguf files (#10094) CI #430: Commit b9e02e8 pushed by ngxson
1h 14m 3s master
master
1h 14m 3s
ggml : add Q4_0_8_8 RISC-V GEMV and GEMM kernels (#10029) CI #429: Commit fc83a9e pushed by ngxson
1h 22m 13s master
master
1h 22m 13s
llama : refactor model loader with backend registry (#10026) CI #428: Commit c5b0f4b pushed by ngxson
1h 11m 50s master
master
1h 11m 50s
ggml: Add POOL2D OP for GPU acceleration to the Vulkan backend in the… CI #427: Commit 8f275a7 pushed by ngxson
1h 19m 14s master
master
1h 19m 14s
llama : Add IBM granite template (#10013) CI #426: Commit 61715d5 pushed by ngxson
1h 0m 23s master
master
1h 0m 23s
musa: workaround for Guilty Lockup in cleaning src0 (#10042) CI #425: Commit 524afee pushed by ngxson
1h 12m 14s master
master
1h 12m 14s
server : don't overfill the batch during infill (#10018) CI #424: Commit 8125e6c pushed by ngxson
1h 10m 43s master
master
1h 10m 43s
llama : switch KQ multiplication to F32 precision by default (#10015) CI #423: Commit 8841ce3 pushed by ngxson
1h 2m 31s master
master
1h 2m 31s
sync : ggml CI #422: Commit cc2983d pushed by ngxson
1h 10m 58s master
master
1h 10m 58s
llama : add DRY sampler (#9702) CI #421: Commit ff252ea pushed by ngxson
1h 7m 50s master
master
1h 7m 50s
llamafile : extend sgemm.cpp support for Q5_0 models (#10010) CI #420: Commit 2f8bd2b pushed by ngxson
1h 14m 23s master
master
1h 14m 23s
server : check that the prompt fits in the slot's context (#10030) CI #419: Commit bc5ba00 pushed by ngxson
1h 25m 26s master
master
1h 25m 26s
server : refactor slot input data, move tokenizer to HTTP thread (#10… CI #418: Commit 958367b pushed by ngxson
1h 10m 47s master
master
1h 10m 47s
CUDA: fix insufficient buffer clearing for MMQ (#10032) CI #417: Commit 167a515 pushed by ngxson
1h 8m 46s master
master
1h 8m 46s
CUDA: fix MMQ for non-contiguous src0, add tests (#10021) CI #416: Commit c39665f pushed by ngxson
1h 9m 53s master
master
1h 9m 53s
server : samplers accept the prompt correctly (#10019) CI #415: Commit 0a1c750 pushed by ngxson
1h 12m 26s master
master
1h 12m 26s