Skip to content

Actions: ggml-org/llama.cpp

Actions

CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
build.yml
6,478 workflow run results
6,478 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

CUDA: fix crash on large batch size for quant. MoE (#13537) CI #22537: Commit 4696d56 pushed by JohannesGaessler
5h 22m 14s master
master
5h 22m 14s
common: add partial regex support CI #22536: Pull request #12808 synchronize by ochafik
3h 19m 56s ochafik:partial-regex
ochafik:partial-regex
3h 19m 56s
llama : fix quantize with dl backends (#13539) CI #22535: Commit b7d2672 pushed by slaren
4h 16m 38s master
master
4h 16m 38s
CUDA: faster Deepseek FA, add Turing support (#13435) CI #22534: Commit 6da34fa pushed by JohannesGaessler
3h 26m 2s master
master
3h 26m 2s
fix: proper error handling for missing elements in messages array (OpenAI compatible backend) CI #22533: Pull request #13540 opened by pwilkin
3h 6m 35s pwilkin:piw-jinjafix
pwilkin:piw-jinjafix
3h 6m 35s
llama : fix quantize with dl backends CI #22532: Pull request #13539 opened by slaren
1h 44m 38s sl/fix-quant-dl
sl/fix-quant-dl
1h 44m 38s
fix: Move build_inp_pos to the top of the graph section for build_gra… CI #22531: Commit 5e7d95e pushed by ggerganov
3h 18m 17s master
master
3h 18m 17s
Granite MoE NoPE fix CI #22530: Pull request #13538 opened by gabe-l-hart
1h 18m 13s gabe-l-hart:GraniteMoENopeFix
gabe-l-hart:GraniteMoENopeFix
1h 18m 13s
server : passthrough the /models endpoint during loading (#13535) CI #22529: Commit 0531744 pushed by ggerganov
1h 45m 14s master
master
1h 45m 14s
server : passthrough the /models endpoint during loading CI #22528: Pull request #13535 synchronize by ggerganov
1h 4m 45s gg/server-models-loading
gg/server-models-loading
1h 4m 45s
CUDA: fix crash on large batch size for quant. MoE CI #22527: Pull request #13537 opened by JohannesGaessler
48m 50s JohannesGaessler:cuda-fix-dsr1-max-batch
JohannesGaessler:cuda-fix-dsr1-max-batch
48m 50s
server : fix cache_tokens bug with no cache_prompt (#13533) CI #22526: Commit 360a9c9 pushed by ngxson
1h 8m 31s master
master
1h 8m 31s
sycl: disable reorder for sycl mulmat CI #22525: Pull request #13536 opened by sgeor255
20m 21s sgeor255:svet/sycl-mulmat-disable-reorder
sgeor255:svet/sycl-mulmat-disable-reorder
20m 21s
server : passthrough the /models endpoint during loading CI #22524: Pull request #13535 opened by ggerganov
52m 5s gg/server-models-loading
gg/server-models-loading
52m 5s
cmake: simplify vulkan shader test logic (#13263) CI #22523: Commit 09d13d9 pushed by bandoti
50m 35s master
master
50m 35s
vulkan: KHR_coopmat flash attention (#13506) CI #22522: Commit 24e86ca pushed by 0cc4m
54m 37s master
master
54m 37s
server : fix cache_tokens bug with no cache_prompt CI #22521: Pull request #13533 opened by ngxson
48m 30s ngxson:xsn/server_no_cache_bug
ngxson:xsn/server_no_cache_bug
48m 30s
ci : upgraded oneAPI version in SYCL workflows and dockerfile CI #22520: Pull request #13532 opened by Alcpz
49m 25s Alcpz:Alcpz/sycl-2025.1.1-update
Alcpz:Alcpz/sycl-2025.1.1-update
49m 25s
sycl: simplify bin_bcast_kernel CI #22519: Pull request #13383 synchronize by AD2605
55m 12s AD2605:ad/simplify_binbcast
AD2605:ad/simplify_binbcast
55m 12s
vulkan: KHR_coopmat flash attention CI #22518: Pull request #13506 synchronize by jeffbolznv
49m 29s jeffbolznv:fa_coopmat1
jeffbolznv:fa_coopmat1
49m 29s
[CUDA backend ONLY] Use just K-cache for MLA + FA: 47% saving on KV-cache size CI #22517: Pull request #13529 opened by jukofyork
53m 38s mla-fa-disable-v-cache
mla-fa-disable-v-cache
53m 38s
[CANN]Support OP MUL_MAT_ID CI #22516: Pull request #13042 synchronize by noemotiovon
54m 14s noemotiovon:opt_support
noemotiovon:opt_support
54m 14s
kv-cache : add SWA support CI #22514: Pull request #13194 synchronize by ggerganov
1h 13m 43s gg/swa
gg/swa
1h 13m 43s
vulkan: workaround FA compile failures on macos (#13517) CI #22513: Commit ab3971f pushed by 0cc4m
1h 2m 54s master
master
1h 2m 54s
[server] accept requests where 'content' is missing entirely, not just null or empty CI #22512: Pull request #13521 opened by andysalerno
53m 44s andysalerno:andysalerno/acceptNullContent
andysalerno:andysalerno/acceptNullContent
53m 44s