Skip to content

Actions: ggml-org/llama.cpp

Actions

CI (AMD)

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
build-amd.yml
105 workflow runs
105 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

SYCL SET operator optimized for F32 tensors (#16350) CI (AMD) #105: Commit ceff6bb pushed by NeoZhangJianyu
In progress master
master
In progress
mtmd : support home-cooked Mistral Small Omni (#14928) CI (AMD) #104: Commit 1bb4f43 pushed by ngxson
44m 48s master
master
44m 48s
sycl : add ARANGE operator (#16362) CI (AMD) #103: Commit b22572e pushed by CISC
46m 8s master
master
46m 8s
CANN: format code using .clang-format (#15863) CI (AMD) #102: Commit 7a50cf3 pushed by hipudding
46m 4s master
master
46m 4s
common : Update the docs on -t --threads (#16236) CI (AMD) #101: Commit 6f5d924 pushed by ggerganov
1h 25m 46s master
master
1h 25m 46s
ggml-cpu: replace putenv with setenv for const-correctness (#16573) CI (AMD) #100: Commit adc9b60 pushed by ggerganov
53m 24s master
master
53m 24s
SYCL: Add GGML_OP_MEAN operator support (#16009) CI (AMD) #99: Commit ee50ee1 pushed by NeoZhangJianyu
47m 49s master
master
47m 49s
cpu : add FLOOR, CEIL, ROUND and TRUNC unary operators (#16083) CI (AMD) #98: Commit 466c191 pushed by CISC
55m 24s master
master
55m 24s
opencl: add q8_0 mm support (#16469) CI (AMD) #97: Commit 0cb7a06 pushed by max-krasnyansky
46m 26s master
master
46m 26s
metal: optimise GGML_OP_SUM (#16559) CI (AMD) #96: Commit f4ce81c pushed by ggerganov
1h 21m 11s master
master
1h 21m 11s
server : fix img token logs (#16595) CI (AMD) #95: Commit 17304cb pushed by ggerganov
47m 1s master
master
47m 1s
llama-quant: add support for mmproj (#16592) CI (AMD) #94: Commit 3e3cb19 pushed by ngxson
46m 17s master
master
46m 17s
CUDA: Changing the CUDA scheduling strategy to spin (#16585) CI (AMD) #93: Commit 5acd455 pushed by ggerganov
46m 41s master
master
46m 41s
server : fix mtmd checkpoints (#16591) CI (AMD) #92: Commit 554fd57 pushed by ngxson
47m 49s master
master
47m 49s
metal : avoid using Metal's gpuAddress property (#16576) CI (AMD) #91: Commit fa882fd pushed by ggerganov
1h 4m 48s master
master
1h 4m 48s
vulkan: Add ACC_TYPE_VEC2 implementation (#16203) CI (AMD) #90: Commit ffa0590 pushed by 0cc4m
45m 33s master
master
45m 33s
CUDA + openCL: fix bug in accessing rms_norm->src while doing fusion … CI (AMD) #89: Commit 120bf70 pushed by slaren
48m 9s master
master
48m 9s
vulkan: Support FA with K/V in F32 (#16543) CI (AMD) #88: Commit 4258e0c pushed by 0cc4m
56m 36s master
master
56m 36s
vulkan: Improve build time for MSVC (#16545) CI (AMD) #87: Commit 7ea15bb pushed by 0cc4m
1h 12m 48s master
master
1h 12m 48s
CUDA: enable FA for FP32 KV cache (#16546) CI (AMD) #86: Commit 9c7185d pushed by JohannesGaessler
55m 56s master
master
55m 56s
CUDA: use fastdiv + ggml_cuda_mad for mmvf (#16557) CI (AMD) #85: Commit 1ee9d0b pushed by JohannesGaessler
1h 20m 55s master
master
1h 20m 55s
CUDA: add fp kernel for larger batch size MoE (#16512) CI (AMD) #84: Commit 48e2fa9 pushed by JohannesGaessler
46m 4s master
master
46m 4s
cuda : remove legacy copy-op pointer indirection code (#16485) CI (AMD) #83: Commit 5b6913c pushed by CISC
52m 22s master
master
52m 22s
server : dynamic token limit for prompt cache (#16560) CI (AMD) #82: Commit bc07349 pushed by ggerganov
46m 14s master
master
46m 14s
graph : support cacheless embeddings with FA and iSWA (#16528) CI (AMD) #81: Commit e38b7c6 pushed by ggerganov
46m 36s master
master
46m 36s