Skip to content

Actions: ngxson/llama.cpp

Actions

Server

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
server.yml
2,138 workflow runs
2,138 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

context : print graph stats for memory-less contexts (#15586) Server #2100: Commit 85cc1ae pushed by ngxson
16m 31s master
master
16m 31s
model : support MiniCPM-V 4.5 (#15575) Server #2099: Commit c4e9239 pushed by ngxson
12m 28s master
master
12m 28s
metal : remove contiguous assertion for src0 in IM2COL (#15577) Server #2098: Commit 0fd90db pushed by ngxson
12m 33s master
master
12m 33s
Add a warning for special devices (#15563) Server #2097: Commit 4c37636 pushed by ngxson
12m 26s master
master
12m 26s
vulkan: Remove splitting for mul_mat_id (#15568) Server #2096: Commit 34bdbbd pushed by ngxson
12m 40s master
master
12m 40s
opencl: fix support ops condition for rms_norm (#15560) Server #2095: Commit f7207b0 pushed by ngxson
13m 37s master
master
13m 37s
vulkan: fix min subgroup 16 condition for mmid subgroup optimization … Server #2094: Commit 4d917cd pushed by ngxson
25m 12s master
master
25m 12s
tests: Generate unique input values for count_equal (#15487) Server #2093: Commit 886b97a pushed by ngxson
15m 34s master
master
15m 34s
metal: fix regression when no metal devices are present (#15531) Server #2092: Commit 111f8d0 pushed by ngxson
16m 55s master
master
16m 55s
model-conversion : set pooling type to none in logits.cpp (#15564) Server #2091: Commit dfd9b5f pushed by ngxson
13m 44s master
master
13m 44s
batched-bench : fix unified KV cache handling + pp timing (#15562) Server #2090: Commit 6b64f74 pushed by ngxson
12m 57s master
master
12m 57s
metal : add FA kernels for HS=40 (#15559) Server #2089: Commit b0ba31f pushed by ngxson
21m 4s master
master
21m 4s
CANN: ROPE cache sin/cos repeat (#15501) Server #2088: Commit c247d06 pushed by ngxson
14m 48s master
master
14m 48s
vulkan: apply MUL_MAT_ID subgroup optimization to non-coopmat devices… Server #2087: Commit 043fb27 pushed by ngxson
15m 12s master
master
15m 12s
vulkan: Support FA with any multiple of 8 head sizes (#15537) Server #2086: Commit c9a24fb pushed by ngxson
12m 27s master
master
12m 27s
vulkan: enable Conv2D for Apple after MoltenVK fixed the bug (#15526) Server #2085: Commit a9c6ffc pushed by ngxson
12m 34s master
master
12m 34s
CUDA: fix half2 -> half conversion for HIP (#15529) Server #2084: Commit 710dfc4 pushed by ngxson
12m 37s master
master
12m 37s
vulkan: optimize rms_norm, and allow the work to spread across multip… Server #2083: Commit 611f419 pushed by ngxson
12m 26s master
master
12m 26s
model : add support for Seed-OSS (#15490) Server #2082: Commit b1afcab pushed by ngxson
15m 7s master
master
15m 7s
chat : fix debug build assertion in trim function (#15520) Server #2081: Commit 21dc4dd pushed by ngxson
13m 41s master
master
13m 41s
vulkan: Rewrite synchronization to allow some overlap between nodes (… Server #2080: Commit 289bf41 pushed by ngxson
14m 16s master
master
14m 16s
vulkan : support ggml_mean (#15393) Server #2079: Commit 0a9b43e pushed by ngxson
12m 35s master
master
12m 35s
test-opt: allow slight inprecision (#15503) Server #2078: Commit e92734d pushed by ngxson
13m 29s master
master
13m 29s
ggml WebGPU: add support for quantization types (#15440) Server #2077: Commit 4536363 pushed by ngxson
16m 7s master
master
16m 7s
model : gpt-oss add response_format support (#15494) Server #2076: Commit 32732f2 pushed by ngxson
13m 45s master
master
13m 45s
ProTip! You can narrow down the results and go further in time using created:<2025-08-22 or the other filters available.