Skip to content

Actions: ynankani/llama.cpp

Actions

CI (android)

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
build-android.yml
20 workflow runs
20 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 … CI (android) #20: Commit d4abd57 pushed by ynankani
18m 41s master
master
18m 41s
jinja: treat a null left operand of in as a plain lookup (#28620) CI (android) #19: Commit b31b71f pushed by ynankani
14m 50s master
master
14m 50s
ggml-cuda: fix divergent barrier in f16 flash attention (#27870) CI (android) #18: Commit b74f590 pushed by ynankani
23m 38s master
master
23m 38s
context : autoscale n_ctx_train when yarn scaling specified (#28030) CI (android) #17: Commit dfc29b6 pushed by ynankani
22m 7s master
master
22m 7s
DeepseekV4: fix rollback with multi-seq (#26756) CI (android) #16: Commit b0539c4 pushed by ynankani
20m 34s master
master
20m 34s
mtmd: support pocket-tts (#26871) CI (android) #15: Commit 6e62ba5 pushed by ynankani
14m 52s master
master
14m 52s
vendor : update cpp-httplib to 0.53.0 (#26821) CI (android) #14: Commit e23e944 pushed by ynankani
44m 32s master
master
44m 32s
CUDA: refactor MMQ kernel configuration (#24127) CI (android) #13: Commit 6eddde0 pushed by ynankani
27m 59s master
master
27m 59s
vulkan: reduce iq1 shared memory usage for mul_mm (#24287) CI (android) #12: Commit d6d0ce8 pushed by ynankani
13m 20s master
master
13m 20s
CUDA: enroll mul_mat_vec_q_moe into pdl (#24087) CI (android) #11: Commit 2154a0f pushed by ynankani
12m 42s master
master
12m 42s
security : disable private disclosures (#23963) CI (android) #10: Commit 02a5701 pushed by ynankani
16m 58s master
master
16m 58s
ci : add missing Linux label to cpu-x64-high-perf runner (#23958) CI (android) #9: Commit e22b0de pushed by ynankani
13m 23s master
master
13m 23s
vulkan: fast path for walsh-hadamard transform (#23687) CI (android) #8: Commit 48e7078 pushed by ynankani
12m 31s master
master
12m 31s
vulkan: Switch MUL_MAT_VEC to 4 K per iteration for F16/32 (#22887) CI (android) #7: Commit c6e4088 pushed by ynankani
10m 59s master
master
10m 59s
scripts : allow wc2wt with an existing branch (#23189) CI (android) #6: Commit c3f95c1 pushed by ynankani
1h 0m 35s master
master
1h 0m 35s
SYCL: fix multi-GPU system RAM exhaustion by using Level Zero allocat… CI (android) #5: Commit 9ed6e19 pushed by ynankani
13m 23s master
master
13m 23s
cuda: fuse snake activation (mul, sin, sqr, mul, add) (#22667) CI (android) #4: Commit 58e68df pushed by ynankani
12m 56s master
master
12m 56s
codeowners : add ZenDNN backend codeowner (#22772) CI (android) #3: Commit 97f06e9 pushed by ynankani
12m 35s master
master
12m 35s
model : remove duplicate wo_s scale after build_attn (Qwen3, LLaMA) (… CI (android) #2: Commit 0f1bb60 pushed by ynankani
12m 49s master
master
12m 49s
common: fix missing exports in llama-common (#22340) CI (android) #1: Commit 5594d13 pushed by ynankani
54m 27s master
master
54m 27s