Skip to content

Actions: PrismML-Eng/llama.cpp

Actions

CI (3rd-party)

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
build-3rd-party.yml
This workflow was disabled manually.
22 workflow runs
22 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
Add Q2_0 quantization: type definition and CPU backend (#24448) CI (3rd-party) #22: Commit bec4772 pushed by khosravipasha
6m 10s master
master
6m 10s
Merge branch 'ggml-org:master' into master CI (3rd-party) #21: Commit 391c775 pushed by khosravipasha
6m 1s master
master
6m 1s
Merge branch 'ggml-org:master' into master CI (3rd-party) #20: Commit 78df1fc pushed by khosravipasha
5m 49s master
master
5m 49s
Hexagon: Process M-tail rows on HMX instead of HVX (#22724) CI (3rd-party) #19: Commit bbeb89d pushed by khosravipasha
5m 28s master
master
5m 28s
pr2wt : symlink .pi (#22386) CI (3rd-party) #18: Commit f535774 pushed by khosravipasha
5m 31s master
master
5m 31s
chat: fix parallel_tool_calls default setting based on model capabili… CI (3rd-party) #17: Commit 8bccdbb pushed by khosravipasha
5m 32s master
master
5m 32s
server: Expose media_tag on /props endpoint. (#22028) CI (3rd-party) #16: Commit 9e5647a pushed by khosravipasha
5m 18s master
master
5m 18s
metal: Implement ROLL op (#21946) CI (3rd-party) #15: Commit ae2d348 pushed by khosravipasha
5m 38s master
master
5m 38s
cuda: Q1_0 initial backend (#21629) CI (3rd-party) #14: Commit 7e72b38 pushed by khosravipasha
5m 35s master
master
5m 35s
docs: listing qwen3-asr and qwen3-omni as supported (#21857) CI (3rd-party) #13: Commit e974923 pushed by khosravipasha
5m 37s master
master
5m 37s
convert : force f16 or f32 on step3-vl conv weights (#21646) CI (3rd-party) #12: Commit 1e9d771 pushed by khosravipasha
5m 14s master
master
5m 14s
CUDA: skip compilation of superfluous FA kernels (#21768) CI (3rd-party) #11: Commit ff5ef82 pushed by khosravipasha
5m 19s master
master
5m 19s
hexagon: add support for linux on snapdragon (#21707) CI (3rd-party) #10: Commit 81069a8 pushed by khosravipasha
5m 28s master
master
5m 28s
vulkan: Support Q1_0 (#21539) CI (3rd-party) #9: Commit 7b69125 pushed by khosravipasha
5m 51s master
master
5m 51s
CUDA: also store node->src->data ptrs for equality check (#21635) CI (3rd-party) #8: Commit d12cc3d pushed by khosravipasha
5m 14s master
master
5m 14s
ggml-webgpu: Add the support of MUL_MAT_ID (#21147) CI (3rd-party) #7: Commit d0a6dfe pushed by khosravipasha
5m 39s master
master
5m 39s
ggml: add Q1_0 1-bit quantization support (CPU) (#21273) CI (3rd-party) #6: Commit 2e1f0a8 pushed by khosravipasha
6m 27s master
master
6m 27s
sync : ggml CI (3rd-party) #5: Commit 6422036 pushed by khosravipasha
5m 34s master
master
5m 34s
sycl : support nvfp4 type in mul_mat (#21227) CI (3rd-party) #4: Commit 6b949d1 pushed by khosravipasha
5m 43s master
master
5m 43s
opencl: add q4_K gemm and gemv kernels for Adreno (#20919) CI (3rd-party) #3: Commit 08f2145 pushed by khosravipasha
5m 25s master
master
5m 25s
models : move the token embedding norms to the first layer (#20943) CI (3rd-party) #2: Commit 9f102a1 pushed by khosravipasha
5m 12s master
master
5m 12s
CANN: support flash attention for head dim not multiple of 16, fix AL… CI (3rd-party) #1: Commit 07ba6d2 pushed by khosravipasha
5m 14s master
master
5m 14s