Skip to content

Actions: ngxson/llama.cpp

Actions

Release

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
release.yml
1,074 workflow runs
1,074 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

common : avoid logging partial messages (which can contain broken UTF… Release #424: Commit 1a67fcc pushed by ngxson
32m 21s master
master
32m 21s
CANN: Add ggml_set_rows (#14943) Release #423: Commit 204f2cf pushed by ngxson
25m 2s master
master
25m 2s
cuda : add softcap fusion (#14907) Release #422: Commit 138b288 pushed by ngxson
19m 21s master
master
19m 21s
CUDA: add roll (#14919) Release #421: Commit 0a5036b pushed by ngxson
18m 4s master
master
18m 4s
test-backend-ops : extend test case filtering (#14865) Release #420: Commit bda6219 pushed by ngxson
50m 38s master
master
50m 38s
llama-bench : use local GPUs along with RPC servers (#14917) Release #419: Commit c556418 pushed by ngxson
57m 53s master
master
57m 53s
ggml-cpu : deduplicate scalar implementations (#14897) Release #418: Commit db16e28 pushed by ngxson
30m 8s master
master
30m 8s
SYCL: Add set_rows support for quantized types (#14883) Release #417: Commit cd1fce6 pushed by ngxson
15m 27s master
master
15m 27s
mtmd : add support for Voxtral (#14862) Release #416: Commit 00fa15f pushed by ngxson
26m 34s master
master
26m 34s
CUDA: fix pointer incrementation in FA (#14916) Release #415: Commit 946b1f6 pushed by ngxson
16m 24s master
master
16m 24s
model : add support for SmallThinker series (#14898) Release #414: Commit 6c6e397 pushed by ngxson
19m 43s master
master
19m 43s
sycl: refactor quantization to q8_1 (#14815) Release #413: Commit afc0e89 pushed by ngxson
27m 25s master
master
27m 25s
vulkan : add fp16 support for the conv_2d kernel (#14872) Release #412: Commit 89d1029 pushed by ngxson
12m 21s master
master
12m 21s
vulkan: skip empty set_rows to avoid invalid API usage (#14860) Release #411: Commit f1a4e72 pushed by ngxson
25m 11s master
master
25m 11s
model : make rope_yarn_log_mul optional for deepseek2 (#14896) Release #410: Commit 4762ad7 pushed by ngxson
20m 8s master
master
20m 8s
llama : fix kq_scale for the attention layers of PLaMo2 (#14892) Release #409: Commit 1dc9614 pushed by ngxson
13m 49s master
master
13m 49s
Docs: add instructions for adding backends (#14889) Release #408: Commit 446595b pushed by ngxson
12m 21s master
master
12m 21s
HIP: Enable Matrix cores for MMQ Kernels, Enable stream-K for CDNA 3 … Release #407: Commit 66906cd pushed by ngxson
1h 3m 17s master
master
1h 3m 17s
CANN: Implement GLU ops (#14884) Release #406: Commit 11dd5a4 pushed by ngxson
26m 18s master
master
26m 18s
musa: fix build warnings (unused variable) (#14869) Release #405: Commit 9b8f3c6 pushed by ngxson
21m 23s master
master
21m 23s
ggml-cpu : disable GGML_NNPA by default due to instability (#14880) Release #404: Commit c7f3169 pushed by ngxson
17m 31s master
master
17m 31s
metal: SSM_SCAN performance (#14743) Release #403: Commit 793c0d7 pushed by ngxson
23m 53s master
master
23m 53s
opencl: add fused rms_norm_mul (#14841) Release #402: Commit ce111d3 pushed by ngxson
15m 36s master
master
15m 36s
ggml : remove invalid portPos specifiers from dot files (#14838) Release #401: Commit e2b7621 pushed by ngxson
55m 59s master
master
55m 59s
context : restore preemptive sched reset when LLAMA_SET_ROWS=0 (#14870) Release #400: Commit c1dbea7 pushed by ngxson
17m 40s master
master
17m 40s
ProTip! You can narrow down the results and go further in time using created:<2025-07-25 or the other filters available.