Skip to content

Actions: ngxson/llama.cpp

Actions

Release

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
release.yml
1,044 workflow runs
1,044 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

graph : fix geglu (#14077) Release #219: Commit 201b31d pushed by ngxson
21m 28s master
master
21m 28s
CANN: Simplify the environment variable setting(#13104) Release #218: Commit e21d2d4 pushed by ngxson
15m 13s master
master
15m 13s
server : fix LRU check (#14079) Release #217: Commit 87d34b3 pushed by ngxson
34m 20s master
master
34m 20s
sycl: Add reorder to Q6_K mmvq implementation (#13885) Release #216: Commit b460d16 pushed by ngxson
35m 6s master
master
35m 6s
add geglu activation function (#14074) Release #215: Commit 91a8ee6 pushed by ngxson
21m 11s master
master
21m 11s
cuda : fix buffer type check with integrated GPUs (#14069) Release #214: Commit 247e5c6 pushed by ngxson
58m 37s master
master
58m 37s
SYCL: Implement few same quantized type copy kernels (#13739) Release #213: Commit 228f34c pushed by ngxson
13m 57s master
master
13m 57s
llama : fix llama_model_chat_template with template name (LLM_KV with… Release #212: Commit 0974ad7 pushed by ngxson
1h 9m 17s master
master
1h 9m 17s
llama : deprecate llama_kv_self_ API (#14030) Release #211: Commit 745aa53 pushed by ngxson
20m 51s master
master
20m 51s
context : fix SWA-related warning for multiple sequences (#14045) Release #210: Commit 487a5e0 pushed by ngxson
26m 26s master
master
26m 26s
llama : support multiple classifier outputs and labels (#13940) Release #209: Commit d17a809 pushed by ngxson
24m 33s master
master
24m 33s
vulkan: Enable VK_KHR_cooperative_matrix extension for Intel Xe2 GPUs… Release #208: Commit 669c13e pushed by ngxson
12m 59s master
master
12m 59s
memory : migrate from llama_kv_cache to more generic llama_memory (#1… Release #207: Commit 7f37b6c pushed by ngxson
22m 35s master
master
22m 35s
llama : allow using mmap without PrefetchVirtualMemory, apply GGML_WI… Release #206: Commit 3a07714 pushed by ngxson
19m 42s master
master
19m 42s
vocab : warn about missing mask token (#14022) Release #205: Commit 9f47fa5 pushed by ngxson
24m 25s master
master
24m 25s
context : fix pos_min initialization upon error decode (#14008) Release #204: Commit 9e31bec pushed by ngxson
13m 55s master
master
13m 55s
vulkan: automatically deduce size of push constants (#13936) Release #203: Commit 5a8ae30 pushed by ngxson
20m 8s master
master
20m 8s
ggml-vulkan: adds support for op CONV_TRANSPOSE_1D (#13813) Release #202: Commit 0d39844 pushed by ngxson
25m 14s master
master
25m 14s
kv-cache : refactor the update/defrag mechanism (#13988) Release #201: Commit 3e63a58 pushed by ngxson
18m 37s master
master
18m 37s
ci : remove cuda 11.7 releases, switch runner to windows 2022 (#13997) Release #200: Commit 2589ad3 pushed by ngxson
41m 53s master
master
41m 53s
releases : use dl backend for linux release, remove arm64 linux relea… Release #199: Commit 4825487 pushed by ngxson
30m 34s master
master
30m 34s
llama-graph : use ggml_repeat_4d (#13998) Release #198: Commit 3ac6753 pushed by ngxson
21m 39s master
master
21m 39s
CUDA: fix FTZ in FA for Gemma 3 (#13991) Release #197: Commit 0b4be4c pushed by ngxson
56m 2s master
master
56m 2s
kv-cache : fix unified::seq_rm to work with seq_id < 0 (#13985) Release #196: Commit e0e806f pushed by ngxson
43m 19s master
master
43m 19s
vulkan: fix warnings in perf logger querypool code (#13937) Release #195: Commit 7e00e60 pushed by ngxson
21m 40s master
master
21m 40s
ProTip! You can narrow down the results and go further in time using created:<2025-06-03 or the other filters available.