Skip to content

Actions: ngxson/llama.cpp

Actions

Release

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
release.yml
1,062 workflow runs
1,062 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

ggml : remove kompute backend (#14501) Release #262: Commit d4cdd9c pushed by ngxson
21m 4s master
master
21m 4s
CUDA: add dynamic shared mem to softmax, refactor general usage (#14497) Release #261: Commit 55c2646 pushed by ngxson
51m 6s master
master
51m 6s
llama : initial Mamba-2 support (#9126) Release #260: Commit 5d46bab pushed by ngxson
1h 14m 49s master
master
1h 14m 49s
Set RPATH to "@loader_path" / "$ORIGIN" to ensure executables and dyn… Release #259: Commit f3ed38d pushed by ngxson
13m 28s master
master
13m 28s
CUDA: add softmax broadcast (#14475) Release #258: Commit 55a1c5a pushed by ngxson
1h 9m 6s master
master
1h 9m 6s
opencl : fix possible buffer overflow in dump_tensor (#14490) Release #257: Commit 307e79d pushed by ngxson
32m 52s master
master
32m 52s
simple-chat : fix context-exceeded condition (#14494) Release #256: Commit d7f5f4e pushed by ngxson
20m 34s master
master
20m 34s
opencl : skip empty nodes on cgraph compute (#14491) Release #255: Commit c8a4e47 pushed by ngxson
23m 39s master
master
23m 39s
opencl : update upscale to support align corners (#14488) Release #254: Commit 603e43d pushed by ngxson
15m 50s master
master
15m 50s
ggml : Callback before abort (#14481) Release #253: Commit 68b3cd6 pushed by ngxson
58m 16s master
master
58m 16s
ci : disable fast-math for Metal GHA CI (#14478) Release #252: Commit de56944 pushed by ngxson
21m 9s master
master
21m 9s
CANN: update aclnnGroupedMatmulV2 to aclnnGroupedMatmulV3 (#14411) Release #251: Commit 343b6e9 pushed by ngxson
1h 31m 18s master
master
1h 31m 18s
vulkan: Split large mul_mat_id to fit in shared memory (#14451) Release #250: Commit 6a746cf pushed by ngxson
1h 12m 7s master
master
1h 12m 7s
add GELU_ERF (#14455) Release #249: Commit eff5e45 pushed by ngxson
1h 30m 59s master
master
1h 30m 59s
ggml : remove trailing whitespace (#0) Release #248: Commit a6a4795 pushed by ngxson
1h 21m 37s master
master
1h 21m 37s
opencl : add GEGLU, REGLU, SWIGLU (#14456) Release #247: Commit 79b33b2 pushed by ngxson
19m 35s master
master
19m 35s
Add Conv2d for CPU (#14388) Release #246: Commit 0a5a3b5 pushed by ngxson
1h 3m 2s master
master
1h 3m 2s
memory : correctly handle failure in apply() (#14438) Release #245: Commit 745f11f pushed by ngxson
14m 33s master
master
14m 33s
metal : disable fast-math for some cpy kernels (#14460) Release #244: Commit 5dd942d pushed by ngxson
21m 27s master
master
21m 27s
ggml-cpu: sycl: Re-enable exp f16 (#14462) Release #243: Commit a7417f5 pushed by ngxson
18m 6s master
master
18m 6s
test-backend-ops : disable llama test (#14461) Release #242: Commit eb3fa29 pushed by ngxson
25m 6s master
master
25m 6s
cmake : Remove redundant include path in CMakeLists.txt (#14452) Release #241: Commit c839a2d pushed by ngxson
15m 55s master
master
15m 55s
server : support jinja extra template kwargs (Qwen3 enable_thinking f… Release #240: Commit caf5681 pushed by ngxson
12m 58s master
master
12m 58s
SYCL: disable faulty fp16 exp kernel (#14395) Release #239: Commit f47c1d7 pushed by ngxson
19m 48s master
master
19m 48s
ggml : fix unmerged GGML_FPxx_TO_FPxx refactoring (#14443) Release #238: Commit a5d1fb6 pushed by ngxson
1h 10m 49s master
master
1h 10m 49s
ProTip! You can narrow down the results and go further in time using created:<2025-06-29 or the other filters available.