Skip to content

Actions: ngxson/llama.cpp

Actions

Server

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
server.yml
2,136 workflow runs
2,136 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

llama : add OLMo November 2024 support (#10394) Server #619: Commit a88ad00 pushed by ngxson
7m 14s master
master
7m 14s
sycl : Add option to set the SYCL architecture for all targets (#10266) Server #618: Commit 2a1507c pushed by ngxson
8m 57s master
master
8m 57s
vulkan: Optimize soft_max (#10301) Server #617: Commit b3e5859 pushed by ngxson
7m 57s master
master
7m 57s
sycl: Revert MUL_MAT_OP support changes (#10385) Server #616: Commit 557924f pushed by ngxson
8m 8s master
master
8m 8s
cuda : only use native when supported by cmake (#10389) Server #615: Commit d3481e6 pushed by ngxson
22m 50s master
master
22m 50s
Vulkan: Fix device info output format specifiers (#10366) Server #614: Commit 9b75f03 pushed by ngxson
18m 33s master
master
18m 33s
CUDA: fix MMV kernel being used for FP16 src1 (#10357) Server #613: Commit 76e9e58 pushed by ngxson
11m 12s master
master
11m 12s
llama : only use default buffer types for the KV cache (#10358) Server #612: Commit be5cacc pushed by ngxson
20m 11s master
master
20m 11s
CUDA: remove DMMV, consolidate F16 mult mat vec (#10318) Server #611: Commit c3ea58a pushed by ngxson
8m 51s master
master
8m 51s
ggml : fix possible buffer use after free in sched reserve (#9930) Server #610: Commit eda7e1d pushed by ngxson
27m 21s master
master
27m 21s
llama/ex: remove --logdir argument (#10339) Server #609: Commit 4e54be0 pushed by ngxson
7m 18s master
master
7m 18s
llamafile : fix include path (#0) Server #608: Commit db4cfd5 pushed by ngxson
7m 57s master
master
7m 57s
server: (web UI) Add samplers sequence customization (#10255) Server #607: Commit bcdb7a2 pushed by ngxson
35m 14s master
master
35m 14s
vulkan: Optimize some mat-vec mul quant shaders (#10296) Server #606: Commit 772703c pushed by ngxson
8m 9s master
master
8m 9s
ggml : optimize Q4_0 into Q4_0_X_Y repack (#10324) Server #605: Commit 1e58ee1 pushed by ngxson
7m 50s master
master
7m 50s
Make updates to fix issues with clang-cl builds while using AVX512 fl… Server #604: Commit 74d73dc pushed by ngxson
8m 36s master
master
8m 36s
scripts: update compare-llama-bench.py (#10319) Server #603: Commit 4047be7 pushed by ngxson
8m 36s master
master
8m 36s
cmake : fix ppc64 check (whisper/0) Server #602: Commit 09ecbcb pushed by ngxson
8m 26s master
master
8m 26s
AVX BF16 and single scale quant optimizations (#10212) Server #601: Commit 1842922 pushed by ngxson
30m 18s master
master
30m 18s
server : (web UI) add copy button for code block, fix api key (#10242) Server #600: Commit 9901068 pushed by ngxson
9m 14s master
master
9m 14s
sycl: Use syclcompat::dp4a (#10267) Server #599: Commit 5a54af4 pushed by ngxson
9m 12s master
master
9m 12s
backend cpu: add online flow for aarch64 Q4_0 GEMV/GEMM kernels (#9921) Server #598: Commit 1607a5e pushed by ngxson
8m 23s master
master
8m 23s
ggml : build backends as libraries (#10256) Server #597: Commit ae8de6d pushed by ngxson
13m 13s master
master
13m 13s
CUDA: no -sm row for very small matrices (#10185) Server #596: Commit 4a8ccb3 pushed by ngxson
22m 36s master
master
22m 36s
speculative : fix out-of-bounds access (#10289) Server #595: Commit 2a82891 pushed by ngxson
20m 1s master
master
20m 1s
ProTip! You can narrow down the results and go further in time using created:<2024-11-14 or the other filters available.