Skip to content

Actions: ggml-org/llama.cpp

Actions

Release

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
release.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
completion : remove useless statics (#24226) Release #3241: Commit 6b80c74 pushed by angt
34m 48s master
master
34m 48s
completion : fix format specifier in LOG_INF (#24213) Release #3240: Commit 588f0dc pushed by angt
34m 20s master
master
34m 20s
mtmd, server: add "placeholder bitmap" for counting tokens , add */in… Release #3239: Commit f5c6ae1 pushed by ngxson
22m 6s master
master
22m 6s
vulkan: check coopmat2 features before reporting support (#24186) Release #3238: Commit 5a69c97 pushed by 0cc4m
25m 50s master
master
25m 50s
model : rename local n_layer_all variable (#24209) Release #3237: Commit 5343f45 pushed by ggerganov
54m 20s master
master
54m 20s
context : fix off-by-one comparisons to n_gpu_layers (#24208) Release #3236: Commit 603300b pushed by ggerganov
31m 28s master
master
31m 28s
opencl: improve get_rows, cpy, concat and q6_k flat gemv (#24160) Release #3235: Commit 308f61c pushed by lhez
27m 35s master
master
27m 35s
common/chat : unify and fix LFM2/LFM2.5 tool parser (#24178) Release #3234: Commit da87e9b pushed by aldehir
38m 4s master
master
38m 4s
vulkan: add fwht support for Intel with shmem reduction (#23964) Release #3233: Commit e82beaa pushed by 0cc4m
35m 56s master
master
35m 56s
model: fix build failed (#24193) Release #3232: Commit c4a278d pushed by ngxson
1h 19m 27s master
master
1h 19m 27s
model, mtmd: Granite4 Vision (#23545) Release #3231: Commit 64086f2 pushed by ngxson
1h 15m 38s master
master
1h 15m 38s
TP: round up granularity to 128 (#24180) Release #3230: Commit 6effcec pushed by JohannesGaessler
1h 15m 50s master
master
1h 15m 50s
cli: fix model params not propagated (#23893) Release #3229: Commit 86591c7 pushed by ngxson
42m 53s master
master
42m 53s
model : fix llama_model::n_gpu_layers() (#24188) Release #3228: Commit 96fbe00 pushed by ggerganov
11m 23s master
master
11m 23s
ui: run npm install when package-lock.json is newer than node_modules… Release #3227: Commit 2016bf2 pushed by ServeurpersoCom
24m 51s master
master
24m 51s
minor : fix lint issues (#24165) Release #3226: Commit 59917d3 pushed by ggerganov
1h 4m 35s master
master
1h 4m 35s
hparams : refactor hparams.n_layer (#24060) Release #3225: Commit 7acb4e8 pushed by ggerganov
39m 15s master
master
39m 15s
kleidiai : dynamic chunck-based scheduling for hybrid execution (#23819) Release #3224: Commit 3ecfb15 pushed by ggerganov
36m 39s master
master
36m 39s
CUDA: enroll mul_mat_vec_q_moe into pdl (#24087) Release #3223: Commit 2154a0f pushed by JohannesGaessler
21m 32s master
master
21m 32s
sycl : port multi-column MMVQ from CUDA backend (#21845) Release #3222: Commit 7fe2ae4 pushed by ggerganov
39m 59s master
master
39m 59s
server : disable on-device spec checkpoints (#24108) Release #3221: Commit 7c158fb pushed by ggerganov
47m 24s master
master
47m 24s
arg: fix double mtp downloads (#24128) Release #3220: Commit 260862b pushed by ggerganov
27m 54s master
master
27m 54s
Move duplicated imatrix code into single common imatrix-loader.cpp (#… Release #3219: Commit e7bcf1c pushed by pwilkin
37m 29s master
master
37m 29s
return filter to save memory (#24125) Release #3218: Commit 0dbfa66 pushed by pwilkin
1h 22m 5s master
master
1h 22m 5s
ggml: vectorize ggml_vec_dot_q4_1_q8_1 with WASM SIMD128 (#22209) Release #3217: Commit 4c51309 pushed by ggerganov
1h 37m 10s master
master
1h 37m 10s