Skip to content

Actions: ggml-org/llama.cpp

Actions

CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
build.yml
362 workflow run results
362 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

sampling : avoid expensive softmax during greedy sampling CI #15458: Pull request #9605 opened by ggerganov
1h 40m 16s gg/sampling-faster-greedy
gg/sampling-faster-greedy
1h 40m 16s
sampling : fix off-by-one in tail-free sampling CI #15457: Pull request #9604 opened by ggerganov
2h 17m 55s gg/tfs-ob1
gg/tfs-ob1
2h 17m 55s
keep the minimum min_keep value to 1 in sampling CI #15456: Pull request #9603 opened by kylo5aby
2h 15m 26s kylo5aby:min_keep_in_samples
kylo5aby:min_keep_in_samples
2h 15m 26s
perplexity : remove extra new lines after chunks (#9596) CI #15455: Commit 37f8c7b pushed by ggerganov
2h 17m 38s master
master
2h 17m 38s
metal : use F32 prec for K*Q in vec FA (#9595) CI #15454: Commit bf9c101 pushed by ggerganov
2h 2m 32s master
master
2h 2m 32s
llama : introduce anonymous namespace in llama.cpp CI #15453: Pull request #9602 opened by danbev
1h 8m 56s danbev:llama-namespace
danbev:llama-namespace
1h 8m 56s
threads: improve ggml_barrier scaling with large number of threads CI #15452: Pull request #9598 synchronize by max-krasnyansky
1h 8m 4s CodeLinaro:improve-ggml-barrier-scaling
CodeLinaro:improve-ggml-barrier-scaling
1h 8m 4s
Revert "[SYCL] fallback mmvq (#9088)" (#9579) CI #15451: Commit e62e978 pushed by airMeng
1h 3m 3s master
master
1h 3m 3s
vocab: refactor tokenizer to reduce the overhead of creating multi times tokenizer CI #15450: Pull request #9449 synchronize by kylo5aby
54m 36s kylo5aby:refactor-tokenizer
kylo5aby:refactor-tokenizer
54m 36s
threads: improve ggml_barrier scaling with large number of threads CI #15449: Pull request #9598 opened by max-krasnyansky
54m 29s CodeLinaro:improve-ggml-barrier-scaling
CodeLinaro:improve-ggml-barrier-scaling
54m 29s
musa: enable VMM support CI #15448: Pull request #9597 opened by yeahdongcn
1h 3m 18s makllama:musa_vmm
makllama:musa_vmm
1h 3m 18s
perplexity : remove extra new lines after chunks CI #15447: Pull request #9596 opened by ggerganov
1h 4m 9s gg/perplexity-nl
gg/perplexity-nl
1h 4m 9s
metal : use F32 prec for K*Q in vec FA CI #15446: Pull request #9595 opened by ggerganov
56m 52s gg/metal-fa-f32-qk
gg/metal-fa-f32-qk
56m 52s
musa: enable building fat binaries, enable unified memory, and disabl… CI #15444: Commit c35e586 pushed by slaren
1h 15m 58s master
master
1h 15m 58s
Fix merge error in #9454 (#9589) CI #15443: Commit 912c331 pushed by slaren
1h 0m 41s master
master
1h 0m 41s
ggml: RWKV_WKV: Fix merge error in #9454 CI #15442: Pull request #9589 opened by MollySophia
54m 9s MollySophia:fix-error
MollySophia:fix-error
54m 9s
musa: enable building fat binaries, enable unified memory, and disable Flash Attention on QY1 (MTT S80) CI #15441: Pull request #9526 synchronize by yeahdongcn
55m 17s makllama:upstream_master
makllama:upstream_master
55m 17s
CUDA: Enable K-shift operation for -ctk q8_0 (limited) CI #15440: Pull request #9571 synchronize by ghost
1h 2m 10s kshift
kshift
1h 2m 10s
CUDA: enable Gemma FA for HIP/Pascal (#9581) CI #15439: Commit a5b57b0 pushed by JohannesGaessler
1h 10m 9s master
master
1h 10m 9s
musa: enable building fat binaries, enable unified memory, and disable Flash Attention on QY1 (MTT S80) CI #15438: Pull request #9526 synchronize by yeahdongcn
55m 46s makllama:upstream_master
makllama:upstream_master
55m 46s
llama: remove redundant loop when constructing ubatch (#9574) CI #15437: Commit ecd5d6b pushed by slaren
2h 1m 12s master
master
2h 1m 12s
RWKV v6: RWKV_WKV op CUDA implementation (#9454) CI #15436: Commit 2a63caa pushed by slaren
1h 46m 34s master
master
1h 46m 34s
RWKV v6: RWKV_WKV op CUDA implementation CI #15435: Pull request #9454 synchronize by MollySophia
51m 11s MollySophia:wkv-cuda
MollySophia:wkv-cuda
51m 11s
CUDA: enable Gemma FA for HIP/Pascal CI #15434: Pull request #9581 opened by JohannesGaessler
1h 3m 25s JohannesGaessler:hip-enable-gemma-fattn
JohannesGaessler:hip-enable-gemma-fattn
1h 3m 25s
Revert "[SYCL] fallback mmvq" CI #15433: Pull request #9579 opened by qnixsynapse
1h 6m 24s qnixsynapse:revert-9088-sycl-fallback-mmvq
qnixsynapse:revert-9088-sycl-fallback-mmvq
1h 6m 24s