Skip to content

Actions: ngxson/llama.cpp

Actions

Server

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
server.yml
1,301 workflow run results
1,301 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

sycl: Add reorder to Q6_K mmvq implementation (#13885) Server #1695: Commit b460d16 pushed by ngxson
28m 51s master
master
28m 51s
add geglu activation function (#14074) Server #1694: Commit 91a8ee6 pushed by ngxson
10m 52s master
master
10m 52s
cuda : fix buffer type check with integrated GPUs (#14069) Server #1693: Commit 247e5c6 pushed by ngxson
11m 28s master
master
11m 28s
SYCL: Implement few same quantized type copy kernels (#13739) Server #1692: Commit 228f34c pushed by ngxson
10m 58s master
master
10m 58s
llama : fix llama_model_chat_template with template name (LLM_KV with… Server #1691: Commit 0974ad7 pushed by ngxson
20m 9s master
master
20m 9s
llama : deprecate llama_kv_self_ API (#14030) Server #1690: Commit 745aa53 pushed by ngxson
10m 17s master
master
10m 17s
context : fix SWA-related warning for multiple sequences (#14045) Server #1689: Commit 487a5e0 pushed by ngxson
11m 20s master
master
11m 20s
llama : support multiple classifier outputs and labels (#13940) Server #1688: Commit d17a809 pushed by ngxson
13m 58s master
master
13m 58s
vulkan: Enable VK_KHR_cooperative_matrix extension for Intel Xe2 GPUs… Server #1687: Commit 669c13e pushed by ngxson
10m 45s master
master
10m 45s
memory : migrate from llama_kv_cache to more generic llama_memory (#1… Server #1686: Commit 7f37b6c pushed by ngxson
10m 42s master
master
10m 42s
llama : allow using mmap without PrefetchVirtualMemory, apply GGML_WI… Server #1685: Commit 3a07714 pushed by ngxson
10m 4s master
master
10m 4s
vocab : warn about missing mask token (#14022) Server #1684: Commit 9f47fa5 pushed by ngxson
11m 24s master
master
11m 24s
context : fix pos_min initialization upon error decode (#14008) Server #1683: Commit 9e31bec pushed by ngxson
11m 59s master
master
11m 59s
vulkan: automatically deduce size of push constants (#13936) Server #1682: Commit 5a8ae30 pushed by ngxson
14m 4s master
master
14m 4s
ggml-vulkan: adds support for op CONV_TRANSPOSE_1D (#13813) Server #1681: Commit 0d39844 pushed by ngxson
16m 52s master
master
16m 52s
kv-cache : refactor the update/defrag mechanism (#13988) Server #1680: Commit 3e63a58 pushed by ngxson
10m 20s master
master
10m 20s
ci : remove cuda 11.7 releases, switch runner to windows 2022 (#13997) Server #1679: Commit 2589ad3 pushed by ngxson
10m 56s master
master
10m 56s
releases : use dl backend for linux release, remove arm64 linux relea… Server #1678: Commit 4825487 pushed by ngxson
11m 19s master
master
11m 19s
llama-graph : use ggml_repeat_4d (#13998) Server #1677: Commit 3ac6753 pushed by ngxson
11m 45s master
master
11m 45s
kv-cache : fix unified::seq_rm to work with seq_id < 0 (#13985) Server #1676: Commit e0e806f pushed by ngxson
10m 17s master
master
10m 17s
vulkan: fix warnings in perf logger querypool code (#13937) Server #1675: Commit 7e00e60 pushed by ngxson
10m 28s master
master
10m 28s
opencl: add backend_synchronize (#13939) Server #1674: Commit 71e74a3 pushed by ngxson
14m 54s master
master
14m 54s
OpenCL: Add concat, tsembd, upscale, tanh, pad and repeat (#13840) Server #1673: Commit bfb1e01 pushed by ngxson
11m 27s master
master
11m 27s
server : disable speculative decoding for SWA models (#13970) Server #1672: Commit 3637576 pushed by ngxson
13m 38s master
master
13m 38s
gemma : more consistent attention scaling for v2 and v3 (#13951) Server #1671: Commit 5582c49 pushed by ngxson
11m 30s master
master
11m 30s