Skip to content

Actions: ngxson/llama.cpp

Actions

Server

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
server.yml
2,500+ workflow runs
2,500+ 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
webui: Server tools (#21237) Server #4399: Commit f42e29f pushed by ngxson
14m 32s master
master
14m 32s
vulkan: add barrier after writetimestamp (#21865) Server #4398: Commit 1982117 pushed by ngxson
14m 11s master
master
14m 11s
ggml : skip already registered backends and devices (#22296) Server #4397: Commit 50494a2 pushed by ngxson
1h 1m 58s master
master
1h 1m 58s
Mirror https://github.com/ggml-org/llama.cpp/pull/22121 Server #4396: Pull request #103 synchronize by julmb
Action required julmb:bounds
julmb:bounds
Action required
CANN: add new ops, optimize existing ops (#21204) Server #4395: Commit c3e08f4 pushed by ngxson
45m 33s master
master
45m 33s
ggml-webgpu: add Q1_0 support (#22374) Server #4394: Commit 434b2a1 pushed by ngxson
42m 9s master
master
42m 9s
server: (router) Forward form-data to model server (Fixes #22044) (#2… Server #4393: Commit 983ca89 pushed by ngxson
18m 33s master
master
18m 33s
(Mirror) Router: Forward form-data to model server Server #4392: Pull request #102 synchronize by tha80
Action required tha80:router-forward-form-data
tha80:router-forward-form-data
Action required
add fast mat-vec kernels for i-quants (#22344) Server #4391: Commit 665abc6 pushed by ngxson
16m 10s master
master
16m 10s
Additional test for common/gemma4 : handle parsing edge cases (#22420) Server #4390: Commit 4414c04 pushed by ngxson
33m 48s master
master
33m 48s
Mirror https://github.com/ggml-org/llama.cpp/pull/22121 Server #4389: Pull request #103 opened by ngxson
16m 34s julmb:bounds
julmb:bounds
16m 34s
Fix type casting for unaccounted memory calculation (#22424) Server #4388: Commit 42401c7 pushed by ngxson
14m 25s master
master
14m 25s
model : remove duplicate wo_s scale after build_attn (Qwen3, LLaMA) (… Server #4387: Commit 0f1bb60 pushed by ngxson
17m 14s master
master
17m 14s
ggml : use 64 bytes aligned tile buffers (#21058) Server #4386: Commit f84270e pushed by ngxson
31m 27s master
master
31m 27s
common: fix missing exports in llama-common (#22340) Server #4385: Commit 5594d13 pushed by ngxson
15m 16s master
master
15m 16s
add performance-portable tuning for register-tile and subgroup matmul… Server #4384: Commit 06a811d pushed by ngxson
37m 52s master
master
37m 52s
Fix recurrent state serialization for partial reads and writes (#22362) Server #4383: Commit 78433f6 pushed by ngxson
28m 43s master
master
28m 43s
CUDA: better coalesce data-access for contiguous concat (#22330) Server #4382: Commit b1a5bd4 pushed by ngxson
16m 11s master
master
16m 11s
ggml-cpu : re-enable fast gelu_quick_f16 (#22339) Server #4381: Commit 0c6ee1c pushed by ngxson
42m 12s master
master
42m 12s
opencl: add iq4_nl support (#22272) Server #4380: Commit f454bd7 pushed by ngxson
16m 6s master
master
16m 6s
hexagon: guard HMX clock request for v75+ platforms (#22377) Server #4379: Commit b760272 pushed by ngxson
19m 6s master
master
19m 6s
chat: fix handling of space in reasoning markers (#22353) Server #4378: Commit dcad77c pushed by ngxson
15m 8s master
master
15m 8s
spec : fix vocab compat checks (#22358) Server #4377: Commit 98dc141 pushed by ngxson
12m 13s master
master
12m 13s
CUDA: reduce MMQ stream-k overhead (#22298) Server #4376: Commit 9725a31 pushed by ngxson
18m 30s master
master
18m 30s
llama-quant : default ftype param Q5_1 --> Q8_0 (#20828) Server #4375: Commit 9d34231 pushed by ngxson
40m 22s master
master
40m 22s