Skip to content

Actions: ZephyrLi-pro/llama.cpp

Actions

Release

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
release.yml
17 workflow runs
17 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
llama-arch: fix DeepSeek4 APE tensor op (#25945) Release #17: Commit 67b9b0e pushed by ZephyrLi-pro
1d 0h 17m 47s master
master
1d 0h 17m 47s
opencl: do not use clCreateBufferWithProperties when targeting CL 2… Release #16: Commit 12127de pushed by ZephyrLi-pro
1d 0h 7m 44s master
master
1d 0h 7m 44s
server: honour per-request reasoning_budget_tokens in chat completion… Release #15: Commit 99f3dc3 pushed by ZephyrLi-pro
1d 0h 15m 51s master
master
1d 0h 15m 51s
ggml-webgpu: tune subgroup split (d_split) in flash_attn_vec (#25418) Release #14: Commit 32e41fa pushed by ZephyrLi-pro
1d 7h 14m 51s master
master
1d 7h 14m 51s
llama-batch: add n_keep_tail in split_equal for recurrent models (#25… Release #13: Commit 230ea9d pushed by ZephyrLi-pro
1d 0h 15m 6s master
master
1d 0h 15m 6s
ui: fake 200 for proxy DELETE req (#25298) Release #12: Commit 898b088 pushed by ZephyrLi-pro
1d 0h 4m 22s master
master
1d 0h 4m 22s
Remove redundant CUDA copies after gated_delta_net. (#23940) Release #11: Commit 5a460de pushed by ZephyrLi-pro
2d 16h 23m 3s master
master
2d 16h 23m 3s
vendor : update cpp-httplib to 0.49.0 (#25218) Release #10: Commit c8ae9a7 pushed by ZephyrLi-pro
1d 16h 38m 9s master
master
1d 16h 38m 9s
llama : add llama_model_ftype_name() (#25134) Release #9: Commit fdb1db8 pushed by ZephyrLi-pro
1d 0h 8m 27s master
master
1d 0h 8m 27s
cuda : prevent integer truncation and overflow errors when using KQ m… Release #8: Commit 0eca4d4 pushed by ZephyrLi-pro
1d 5h 0m 40s master
master
1d 5h 0m 40s
Revert "sched : reintroduce less synchronizations during split comput… Release #7: Commit 86b9470 pushed by ZephyrLi-pro
1d 0h 10m 44s master
master
1d 0h 10m 44s
CUDA: batch out_prod broadcast (dps2>1) path with cublasSgemmBatched … Release #6: Commit f818065 pushed by ZephyrLi-pro
1d 0h 6m 0s master
master
1d 0h 6m 0s
server : Add id to tool call responses api (#24882) Release #5: Commit dec5ca5 pushed by ZephyrLi-pro
1d 0h 10m 24s master
master
1d 0h 10m 24s
wasm : fix fallback symbol collision (#24639) Release #4: Commit 6eab471 pushed by ZephyrLi-pro
1d 0h 0m 51s master
master
1d 0h 0m 51s
vendor : update LibreSSL to 4.3.2 (#24397) Release #3: Commit ac4cdde pushed by ZephyrLi-pro
1d 0h 4m 31s master
master
1d 0h 4m 31s
ggml-webgpu: Improve prefill speeds for k-quants + refactor matmul fo… Release #2: Commit 1e1aca0 pushed by ZephyrLi-pro
1d 0h 9m 7s master
master
1d 0h 9m 7s
mtmd : support MiniCPM-V 4.6 (#22529) Release #1: Commit 2496f9c pushed by ZephyrLi-pro
1d 0h 0m 2s master
master
1d 0h 0m 2s