Skip to content

Actions: ggml-org/llama.cpp

Actions

CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
build.yml
14,677 workflow runs
14,677 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

model : add ASR support for LFM2-Audio-1.5B (conformer) (#18106) CI #32581: Commit 8ea958d pushed by ngxson
Queued master
master
Queued
vulkan: Warptile tuning for Intel Xe2/Xe3 CI #32580: Pull request #18178 synchronize by virajwad
Action required virajwad:xe3_perf_changes
virajwad:xe3_perf_changes
Action required
model : add ASR support for LFM2-Audio-1.5B (conformer) CI #32579: Pull request #18106 synchronize by ngxson
Queued tarek/feat/lfm2-asr-upstream
tarek/feat/lfm2-asr-upstream
Queued
model : add ASR support for LFM2-Audio-1.5B (conformer) CI #32578: Pull request #18106 synchronize by ngxson
5m 30s tarek/feat/lfm2-asr-upstream
tarek/feat/lfm2-asr-upstream
5m 30s
presets: refactor, allow cascade presets from different sources, add global section CI #32577: Pull request #18169 synchronize by ngxson
Queued ngxson:xsn/refactor_server_preset
ngxson:xsn/refactor_server_preset
Queued
presets: refactor, allow cascade presets from different sources, add global section CI #32576: Pull request #18169 synchronize by ngxson
25m 46s ngxson:xsn/refactor_server_preset
ngxson:xsn/refactor_server_preset
25m 46s
cmake: Added more x86_64 CPU backends when building with GGML_CPU_ALL_VARIANTS=On CI #32575: Pull request #18186 opened by bberberov
Queued bberberov:more-cpu-backends
bberberov:more-cpu-backends
Queued
implement Power Law sampling CI #32574: Pull request #17927 synchronize by ddh0
3h 7m 4s ddh0:power-law-sampler
ddh0:power-law-sampler
3h 7m 4s
ggml-hexagon: Implement true Q8_0 quantization on Hexagon NPU for more accurate mixed-precision matmul operations CI #32573: Pull request #17977 synchronize by ngdxzy
Action required ngdxzy:real_q8_0
ngdxzy:real_q8_0
Action required
Adding --direct-io flag for model loading CI #32572: Pull request #18166 synchronize by JTischbein
Queued JTischbein:direct_io_flag
JTischbein:direct_io_flag
Queued
ggml-hexagon: gelu optimization CI #32571: Pull request #18151 synchronize by joeldushouyu
3h 4m 44s joeldushouyu:hexagon-gelu-optimization
joeldushouyu:hexagon-gelu-optimization
3h 4m 44s
Adding --direct-io flag for model loading CI #32570: Pull request #18166 synchronize by JTischbein
55m 13s JTischbein:direct_io_flag
JTischbein:direct_io_flag
55m 13s
vulkan: fix im2col overflowing maxworkgroupcount CI #32569: Pull request #18180 opened by jeffbolznv
6h 19m 19s jeffbolznv:im2col_wglimit
jeffbolznv:im2col_wglimit
6h 19m 19s
[DRAFT] CUDA: Improve performance via less synchronizations between token CI #32568: Pull request #17795 synchronize by aendk
4h 58m 57s aendk:akieslinger/reduce-per-token-syncs
aendk:akieslinger/reduce-per-token-syncs
4h 58m 57s
ggml-hexagon: Implement true Q8_0 quantization on Hexagon NPU for more accurate mixed-precision matmul operations CI #32567: Pull request #17977 synchronize by ngdxzy
3h 13m 54s ngdxzy:real_q8_0
ngdxzy:real_q8_0
3h 13m 54s
vulkan: Warptile tuning for Intel Xe2/Xe3 CI #32566: Pull request #18178 opened by virajwad
Action required virajwad:xe3_perf_changes
virajwad:xe3_perf_changes
Action required
llama-server: friendlier error msg when ctx < input CI #32565: Pull request #18174 synchronize by am17an
4h 32m 26s am17an:server-better-error-msg
am17an:server-better-error-msg
4h 32m 26s
presets: refactor, allow cascade presets from different sources, add global section CI #32564: Pull request #18169 synchronize by ngxson
3h 28m 3s ngxson:xsn/refactor_server_preset
ngxson:xsn/refactor_server_preset
3h 28m 3s
[DRAFT] CUDA: Improve performance via less synchronizations between token CI #32563: Pull request #17795 synchronize by aendk
56m 34s aendk:akieslinger/reduce-per-token-syncs
aendk:akieslinger/reduce-per-token-syncs
56m 34s
ggml-hexagon: gelu optimization CI #32562: Pull request #18151 synchronize by joeldushouyu
3h 32m 27s joeldushouyu:hexagon-gelu-optimization
joeldushouyu:hexagon-gelu-optimization
3h 32m 27s
llama-server: friendlier error msg when ctx < input CI #32561: Pull request #18174 opened by am17an
1h 35m 59s am17an:server-better-error-msg
am17an:server-better-error-msg
1h 35m 59s
presets: refactor, allow cascade presets from different sources, add global section CI #32560: Pull request #18169 synchronize by ngxson
2h 17m 39s ngxson:xsn/refactor_server_preset
ngxson:xsn/refactor_server_preset
2h 17m 39s
ggml-cpu: extend support for RVV floating-point kernels (#17318) CI #32559: Commit f716588 pushed by ggerganov
9h 38m 49s master
master
9h 38m 49s
arg: fix ASAN error on sampler_type_names empty (#18167) CI #32558: Commit 4d1316c pushed by ngxson
9h 19m 9s master
master
9h 19m 9s
presets: refactor, allow cascade presets from different sources, add global section CI #32557: Pull request #18169 synchronize by ngxson
1h 8m 44s ngxson:xsn/refactor_server_preset
ngxson:xsn/refactor_server_preset
1h 8m 44s