Skip to content

Actions: ngxson/llama.cpp

Actions

CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
build.yml
1,742 workflow run results
1,742 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

opencl: fix support ops condition for rms_norm (#15560) CI #2001: Commit f7207b0 pushed by ngxson
1h 10m 4s master
master
1h 10m 4s
vulkan: fix min subgroup 16 condition for mmid subgroup optimization … CI #2000: Commit 4d917cd pushed by ngxson
1h 27m 20s master
master
1h 27m 20s
tests: Generate unique input values for count_equal (#15487) CI #1999: Commit 886b97a pushed by ngxson
1h 9m 1s master
master
1h 9m 1s
metal: fix regression when no metal devices are present (#15531) CI #1998: Commit 111f8d0 pushed by ngxson
1h 0m 5s master
master
1h 0m 5s
model-conversion : set pooling type to none in logits.cpp (#15564) CI #1997: Commit dfd9b5f pushed by ngxson
57m 4s master
master
57m 4s
batched-bench : fix unified KV cache handling + pp timing (#15562) CI #1996: Commit 6b64f74 pushed by ngxson
54m 40s master
master
54m 40s
metal : add FA kernels for HS=40 (#15559) CI #1995: Commit b0ba31f pushed by ngxson
57m 33s master
master
57m 33s
CANN: ROPE cache sin/cos repeat (#15501) CI #1994: Commit c247d06 pushed by ngxson
53m 47s master
master
53m 47s
vulkan: apply MUL_MAT_ID subgroup optimization to non-coopmat devices… CI #1993: Commit 043fb27 pushed by ngxson
55m 15s master
master
55m 15s
vulkan: Support FA with any multiple of 8 head sizes (#15537) CI #1992: Commit c9a24fb pushed by ngxson
1h 7m 39s master
master
1h 7m 39s
vulkan: enable Conv2D for Apple after MoltenVK fixed the bug (#15526) CI #1991: Commit a9c6ffc pushed by ngxson
58m 39s master
master
58m 39s
CUDA: fix half2 -> half conversion for HIP (#15529) CI #1990: Commit 710dfc4 pushed by ngxson
1h 3m 7s master
master
1h 3m 7s
vulkan: optimize rms_norm, and allow the work to spread across multip… CI #1989: Commit 611f419 pushed by ngxson
1h 1m 46s master
master
1h 1m 46s
model : add support for Seed-OSS (#15490) CI #1988: Commit b1afcab pushed by ngxson
1h 7m 18s master
master
1h 7m 18s
chat : fix debug build assertion in trim function (#15520) CI #1987: Commit 21dc4dd pushed by ngxson
58m 43s master
master
58m 43s
vulkan: Rewrite synchronization to allow some overlap between nodes (… CI #1986: Commit 289bf41 pushed by ngxson
1h 11m 11s master
master
1h 11m 11s
vulkan : support ggml_mean (#15393) CI #1985: Commit 0a9b43e pushed by ngxson
1h 9m 22s master
master
1h 9m 22s
test-opt: allow slight inprecision (#15503) CI #1984: Commit e92734d pushed by ngxson
1h 9m 24s master
master
1h 9m 24s
ggml WebGPU: add support for quantization types (#15440) CI #1983: Commit 4536363 pushed by ngxson
1h 4m 42s master
master
1h 4m 42s
model : gpt-oss add response_format support (#15494) CI #1982: Commit 32732f2 pushed by ngxson
1h 7m 55s master
master
1h 7m 55s
ggml: add conv3d op (#15182) CI #1981: Commit 92f7f0a pushed by ngxson
1h 3m 7s master
master
1h 3m 7s
cuda : add Pad Reflect 1D support (#14659) CI #1980: Commit b1ab918 pushed by ngxson
58m 6s master
master
58m 6s
llama : remove KV cache defragmentation logic (#15473) CI #1979: Commit 9ebebef pushed by ngxson
1h 6m 32s master
master
1h 6m 32s
ggml-cpu: Support Q5_0 and Q5_1 on s390x (#15486) CI #1978: Commit ad5c975 pushed by ngxson
57m 47s master
master
57m 47s
readme : model : mtdm : lfm2 improvements (#15476) CI #1977: Commit e288693 pushed by ngxson
1h 2m 31s master
master
1h 2m 31s