Skip to content

Actions: qualcomm/llama.cpp

Actions

Check vendor

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
check-vendor.yml
29 workflow runs
29 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
opencl: extend the elementwise and data‐movement op coverage (#27633) Check vendor #29: Commit 1548a24 pushed by lhez
1d 0h 0m 2s master
master
1d 0h 0m 2s
common, server : enable preserve_reasoning kwarg by default, log its … Check vendor #28: Commit e750b88 pushed by max-krasnyansky
1d 0h 0m 2s master
master
1d 0h 0m 2s
chat : add new template for DeepSeek V4 Flash 0731 (#26398) Check vendor #27: Commit 0ef6e55 pushed by tboinovski1
1d 0h 0m 11s master
master
1d 0h 0m 11s
ui: reduce per-token render cost when streaming (#26053) Check vendor #26: Commit 555881e pushed by lhez
1d 0h 0m 2s master
master
1d 0h 0m 2s
llama_dsv4: write only used rows in state (#25325) Check vendor #25: Commit 91d2fc3 pushed by tboinovski1
1d 0h 0m 1s master
master
1d 0h 0m 1s
vulkan: Use native e2m1 and e4m3 conversions for mxfp4/nvfp4 (#25338) Check vendor #24: Commit e920c52 pushed by lhez
1d 0h 0m 1s master
master
1d 0h 0m 1s
Add Q2_0 quantization: type definition and CPU backend (#24448) Check vendor #23: Commit bec4772 pushed by lhez
1d 0h 0m 1s master
master
1d 0h 0m 1s
mtmd: model: unlimited-ocr: converter + parity test (#24969) Check vendor #22: Commit 894bb27 pushed by tboinovski1
1d 0h 0m 2s master
master
1d 0h 0m 2s
chat : fix LFM2 tool-call parsing double-escaping (#24667) Check vendor #21: Commit 7dad2f1 pushed by tboinovski1
1d 0h 0m 2s master
master
1d 0h 0m 2s
server : disable on-device spec checkpoints (#24108) Check vendor #20: Commit 7c158fb pushed by tboinovski1
1d 0h 0m 2s master
master
1d 0h 0m 2s
speculative : fix n_outputs_max and remove draft-simple auto-enable (… Check vendor #19: Commit 5dcb711 pushed by tboinovski1
1d 0h 0m 1s master
master
1d 0h 0m 1s
server : handle If-None-Match weak ETags (#23916) Check vendor #18: Commit 6f165c1 pushed by tboinovski1
1d 0h 0m 2s master
master
1d 0h 0m 2s
metal : add apple device id (#23566) Check vendor #17: Commit 35c9b1f pushed by tboinovski1
1d 0h 0m 2s master
master
1d 0h 0m 2s
hexagon: add support for TRI op (#22822) Check vendor #16: Commit 9a532ae pushed by tboinovski1
1m 13s master
master
1m 13s
Feature hexagon cpy fast path Check vendor #14: Pull request #6 opened by pdhinaka
1m 12s feature-hexagon-cpy-fast-path
feature-hexagon-cpy-fast-path
1m 12s
Hexagon: Process M-tail rows on HMX instead of HVX (#22724) Check vendor #13: Commit bbeb89d pushed by tboinovski1
1m 17s master
master
1m 17s
ggml-webgpu: Improve performance of mat-vec and mat-mat for MUL_MAT_I… Check vendor #12: Commit a95a11e pushed by tboinovski1
1m 8s master
master
1m 8s
parser: fix structured output bug (#22302) Check vendor #11: Commit 0adede8 pushed by tboinovski1
52s master
master
52s
[WebGPU] Implement async tensor api and event api (#22099) Check vendor #10: Commit 0d0764d pushed by tboinovski1
52s master
master
52s
ggml-cuda: flush legacy pool on OOM and retry (#22155) Check vendor #9: Commit 9789512 pushed by max-krasnyansky
51s master
master
51s
ggml : remove ggml-ext.h (#21869) Check vendor #8: Commit fae3a28 pushed by max-krasnyansky
1m 4s master
master
1m 4s
fix **/x glob matching (#21129) Check vendor #7: Commit 6509718 pushed by max-krasnyansky
55s master
master
55s
opencl: add q6_K gemm and gemv kernels for Adreno (#20089) Check vendor #6: Commit 1772701 pushed by max-krasnyansky
55s master
master
55s
convert : fix/suppress pyright errors (#20442) Check vendor #5: Commit 983df14 pushed by lhez
53s master
master
53s