Skip to content

Actions: ngxson/llama.cpp

Actions

Publish Docker image

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
docker.yml
181 workflow run results
181 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

llama : fix buffer checks for mamba and rwk (#10111) Publish Docker image #433: Commit c02e5ab pushed by ngxson
30m 46s master
master
30m 46s
loader: refactor tensor weights storage (#9935) Publish Docker image #432: Commit ab3d71f pushed by ngxson
30m 9s master
master
30m 9s
server : include scheme when printing URL (#10106) Publish Docker image #431: Commit 0a683e8 pushed by ngxson
45m 58s master
master
45m 58s
ggml : check tensor name lengths in gguf files (#10100) Publish Docker image #430: Commit dea5e86 pushed by ngxson
33m 4s master
master
33m 4s
kompute: add mul_mat_q4_k shader (#10097) Publish Docker image #429: Commit 1329c0a pushed by ngxson
30m 40s master
master
30m 40s
kompute: add backend registry / device interfaces (#10045) Publish Docker image #428: Commit 61408e7 pushed by ngxson
31m 59s master
master
31m 59s
ggml : fix memory leaks when loading invalid gguf files (#10094) Publish Docker image #427: Commit b9e02e8 pushed by ngxson
30m 52s master
master
30m 52s
ggml : add Q4_0_8_8 RISC-V GEMV and GEMM kernels (#10029) Publish Docker image #426: Commit fc83a9e pushed by ngxson
1h 52m 47s master
master
1h 52m 47s
llama : refactor model loader with backend registry (#10026) Publish Docker image #425: Commit c5b0f4b pushed by ngxson
32m 5s master
master
32m 5s
ggml: Add POOL2D OP for GPU acceleration to the Vulkan backend in the… Publish Docker image #424: Commit 8f275a7 pushed by ngxson
32m 27s master
master
32m 27s
llama : Add IBM granite template (#10013) Publish Docker image #423: Commit 61715d5 pushed by ngxson
42m 14s master
master
42m 14s
musa: workaround for Guilty Lockup in cleaning src0 (#10042) Publish Docker image #422: Commit 524afee pushed by ngxson
30m 35s master
master
30m 35s
server : don't overfill the batch during infill (#10018) Publish Docker image #421: Commit 8125e6c pushed by ngxson
32m 26s master
master
32m 26s
llama : switch KQ multiplication to F32 precision by default (#10015) Publish Docker image #420: Commit 8841ce3 pushed by ngxson
5m 55s master
master
5m 55s
sync : ggml Publish Docker image #419: Commit cc2983d pushed by ngxson
32m 54s master
master
32m 54s
llama : add DRY sampler (#9702) Publish Docker image #418: Commit ff252ea pushed by ngxson
47m 20s master
master
47m 20s
llamafile : extend sgemm.cpp support for Q5_0 models (#10010) Publish Docker image #417: Commit 2f8bd2b pushed by ngxson
36m 45s master
master
36m 45s
server : check that the prompt fits in the slot's context (#10030) Publish Docker image #416: Commit bc5ba00 pushed by ngxson
34m 46s master
master
34m 46s
server : refactor slot input data, move tokenizer to HTTP thread (#10… Publish Docker image #415: Commit 958367b pushed by ngxson
42m 38s master
master
42m 38s
CUDA: fix insufficient buffer clearing for MMQ (#10032) Publish Docker image #414: Commit 167a515 pushed by ngxson
32m 25s master
master
32m 25s
CUDA: fix MMQ for non-contiguous src0, add tests (#10021) Publish Docker image #413: Commit c39665f pushed by ngxson
30m 51s master
master
30m 51s
server : samplers accept the prompt correctly (#10019) Publish Docker image #412: Commit 0a1c750 pushed by ngxson
30m 49s master
master
30m 49s
llama.vim : add classic vim support (#9995) Publish Docker image #411: Commit ac113a0 pushed by ngxson
32m 40s master
master
32m 40s
llama : fix empty batch causing llama_batch_allocr to crash (#9966) Publish Docker image #410: Commit c8c07d6 pushed by ngxson
30m 57s master
master
30m 57s
llama : rename batch to ubatch (#9950) Publish Docker image #409: Commit 19d900a pushed by ngxson
31m 14s master
master
31m 14s