Skip to content

Actions: ngxson/llama.cpp

Actions

Release

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
release.yml
968 workflow runs
968 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

server : do not return error out of context (with ctx shift disabled)… Release #68: Commit 6aa892e pushed by ngxson
28m 6s master
master
28m 6s
releases : use arm version of curl for arm releases (#13592) Release #67: Commit 415e40a pushed by ngxson
30m 51s master
master
30m 51s
metal : add FA-vec kernel for head size 64 (#13583) Release #66: Commit 654a677 pushed by ngxson
25m 46s master
master
25m 46s
llama : print hint when loading a model when no backends are loaded (… Release #65: Commit 5364ae4 pushed by ngxson
27m 21s master
master
27m 21s
sycl : fixed compilation warnings (#13582) Release #64: Commit 0a338ed pushed by ngxson
33m 6s master
master
33m 6s
minja: sync (qwen3) (#13573) Release #63: Commit bc098c3 pushed by ngxson
50m 11s master
master
50m 11s
gguf : use ggml log system (#13571) Release #62: Commit c6a2c9e pushed by ngxson
20m 2s master
master
20m 2s
convert : fix conversion for llama 4 (#13567) Release #61: Commit c531edf pushed by ngxson
24m 2s master
master
24m 2s
sycl: use oneDNN for matrices multiplication (#12972) Release #60: Commit 9c404ed pushed by ngxson
24m 33s master
master
24m 33s
llama-bench : fix -ot with dl backends (#13563) Release #59: Commit 6c8b915 pushed by ngxson
13m 41s master
master
13m 41s
server : proper error handling for missing elements in messages array… Release #58: Commit c753d7b pushed by ngxson
30m 33s master
master
30m 33s
bench : handle decode errors (#13548) Release #57: Commit b283804 pushed by ngxson
25m 35s master
master
25m 35s
server: inject date_string in llama 3.x template + fix date for fir… Release #56: Commit aa48e37 pushed by ngxson
33m 53s master
master
33m 53s
arm64: optimize q6_k_q8_k kernel with i8mm (#13519) Release #55: Commit 5ab5d5f pushed by ngxson
17m 34s master
master
17m 34s
common: add partial regex support (#12808) Release #54: Commit 3198405 pushed by ngxson
27m 50s master
master
27m 50s
editorconfig : fix trailing whitespace from #13542 (#13546) Release #53: Commit f5170c1 pushed by ngxson
49m 7s master
master
49m 7s
fix: crash when calling llama_state_get_size on a context without a… Release #52: Commit 017f10b pushed by ngxson
27m 45s master
master
27m 45s
CUDA: fix crash on large batch size for quant. MoE (#13537) Release #51: Commit 4696d56 pushed by ngxson
35m 4s master
master
35m 4s
llama : fix quantize with dl backends (#13539) Release #50: Commit b7d2672 pushed by ngxson
49m 32s master
master
49m 32s
CUDA: faster Deepseek FA, add Turing support (#13435) Release #49: Commit 6da34fa pushed by ngxson
41m 34s master
master
41m 34s
fix: Move build_inp_pos to the top of the graph section for build_gra… Release #48: Commit 5e7d95e pushed by ngxson
40m 13s master
master
40m 13s
server : passthrough the /models endpoint during loading (#13535) Release #47: Commit 0531744 pushed by ngxson
38m 37s master
master
38m 37s
server : fix cache_tokens bug with no cache_prompt (#13533) Release #46: Commit 360a9c9 pushed by ngxson
25m 41s master
master
25m 41s
cmake: simplify vulkan shader test logic (#13263) Release #45: Commit 09d13d9 pushed by ngxson
33m 11s master
master
33m 11s
vulkan: KHR_coopmat flash attention (#13506) Release #44: Commit 24e86ca pushed by ngxson
29m 24s master
master
29m 24s
ProTip! You can narrow down the results and go further in time using created:<2025-05-14 or the other filters available.