Skip to content

Actions: ngxson/llama.cpp

Actions

Server

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
server.yml
1,431 workflow run results
1,431 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

gemma : more consistent attention scaling for v2 and v3 (#13951) Server #1671: Commit 5582c49 pushed by ngxson
11m 30s master
master
11m 30s
server: update deepseek reasoning format (pass reasoning_content as… Server #1670: Commit c9bbc77 pushed by ngxson
12m 23s master
master
12m 23s
mtmd : fix memory leak in mtmd_helper_eval_chunk_single (#13961) Server #1669: Commit bfd3227 pushed by ngxson
10m 38s master
master
10m 38s
cmake : Handle mixed-case 'Power' strings in POWER CPU detection (#13… Server #1668: Commit 093e3f1 pushed by ngxson
11m 5s master
master
11m 5s
sycl: quantize and reorder the input to q8_1 when reorder is enabled … Server #1667: Commit 663445b pushed by ngxson
10m 6s master
master
10m 6s
gguf: fix failure on version == 0 (#13956) Server #1666: Commit 7675c55 pushed by ngxson
17m 43s master
master
17m 43s
convert : fix nomic-bert-moe mask token (#13757) Server #1665: Commit 5e1c3ae pushed by ngxson
21m 9s master
master
21m 9s
ggml: check if non-native endian model is being loaded (#13943) Server #1664: Commit e57bb87 pushed by ngxson
11m 2s master
master
11m 2s
sync : ggml Server #1663: Commit f3a4b16 pushed by ngxson
10m 49s master
master
10m 49s
parallel : fix n_junk == 0 (#13952) Server #1662: Commit c046217 pushed by ngxson
10m 26s master
master
10m 26s
threading: support for GGML_SCHED_PRIO_LOW, update thread info on Win… Server #1661: Commit 053b153 pushed by ngxson
20m 17s master
master
20m 17s
server: allow unclosed thinking tags (#13931) Server #1660: Commit e15898d pushed by ngxson
13m 30s master
master
13m 30s
llama : deprecate explicit kv_self defrag/update calls (#13921) Server #1659: Commit 803f8ba pushed by ngxson
22m 49s master
master
22m 49s
llama : use n_swa + n_ubatch cells for SWA cache (#13833) Server #1658: Commit 3600cc2 pushed by ngxson
14m 17s master
master
14m 17s
webui : Replace alert and confirm with custom modals. (#13711) Server #1657: Commit c7e0a20 pushed by ngxson
15m 44s master
master
15m 44s
mtmd : drop _shared from libmtmd name, merge helpers into libmtmd… Server #1656: Commit 51fa76f pushed by ngxson
11m 36s master
master
11m 36s
CUDA: add a prop in ggml_cuda_device_infor for distinguish iGPU or dG… Server #1655: Commit eb39499 pushed by ngxson
11m 28s master
master
11m 28s
sched : avoid changing cur_copy when a graph is already allocated (#1… Server #1654: Commit b47ab7b pushed by ngxson
10m 31s master
master
10m 31s
parallel : increase the variability of the prompt lengths (#13927) Server #1653: Commit dd665cc pushed by ngxson
10m 20s master
master
10m 20s
cuda : prevent using split buffers with 3d/4d matrices (#13919) Server #1652: Commit df0c0c7 pushed by ngxson
15m 51s master
master
15m 51s
SYCL: Add mrope kernel (#13755) Server #1651: Commit b49a8ff pushed by ngxson
11m 28s master
master
11m 28s
sync : vendor (#13901) Server #1650: Commit 53f9250 pushed by ngxson
12m 32s master
master
12m 32s
convert : allow partial update to the chkhsh pre-tokenizer list (#13847) Server #1649: Commit 07e4351 pushed by ngxson
11m 20s master
master
11m 20s
llama : add support for DistilBert (#13907) Server #1648: Commit 291f2b6 pushed by ngxson
11m 16s master
master
11m 16s
llama : use llm_build_granite for minicpm (#13911) Server #1647: Commit 2c90da4 pushed by ngxson
11m 1s master
master
11m 1s
ProTip! You can narrow down the results and go further in time using created:<2025-05-30 or the other filters available.