Default Branch

ebd048fc5e · opencl: flash attention improvement (#25069) · Updated 2026-06-27 17:36:06 -05:00

Branches

87f18f760e · ci : add self-hosted ui workflow · Updated 2026-05-24 14:18:31 -05:00    jdelony

521
10

16b648c897 · ci : try ui SH · Updated 2026-05-24 13:31:25 -05:00    jdelony

521
10

36aa88a853 · cont : move e2e to SH · Updated 2026-05-24 12:00:15 -05:00    jdelony

521
9

ced88c03cb · ci : remove tag from build-self-hosted.yml · Updated 2026-05-24 10:08:11 -05:00    jdelony

522
2

f36e2ab022 · add reverse order tests for dmabuf · Updated 2026-05-21 07:44:52 -05:00    jdelony

554
9

aa27b85ecf · metal : optimize pad · Updated 2026-05-19 12:17:14 -05:00    jdelony

593
1

938872e93f · fix partial writes · Updated 2026-05-15 09:00:57 -05:00    jdelony

667
10

6eb6d84e46 · metal: add GDN partial rollback · Updated 2026-05-14 02:24:09 -05:00    jdelony

697
12

c8f8e2364c · cont : simplify · Updated 2026-05-11 02:54:07 -05:00    jdelony

764
38

efa2f8e5a7 · naming : improve consistency · Updated 2026-05-08 04:24:57 -05:00    jdelony

764
24

ba72d4d287 · ggml: update SCHED_DEBUG output to use ggml_op_desc() · Updated 2026-05-07 18:52:20 -05:00    jdelony

761
1

0445829c1d · llama : enable layer input extraction · Updated 2026-05-05 12:50:20 -05:00    jdelony

791
1

f84632951a · wip · Updated 2026-05-05 01:36:07 -05:00    jdelony

799
23

82af405161 · arg : silence warnings about removed params · Updated 2026-05-04 02:07:57 -05:00    jdelony

811
1

81eabb4781 · sync : ggml · Updated 2026-05-02 00:53:10 -05:00    jdelony

826
2

9d5887035f · testing · Updated 2026-04-30 11:18:57 -05:00    jdelony

840
2

6eddb1c6e3 · pi : add rule to use gh CLI for GitHub resources · Updated 2026-04-30 01:49:54 -05:00    jdelony

843
2

c6a04cb5c3 · ggml-metal: fix 2D async copy to use row-by-row transfers · Updated 2026-04-29 06:57:48 -05:00    jdelony

853
3

fd6f79c7a4 · download : prefer q8_0 when q4_k not available · Updated 2026-04-27 04:08:25 -05:00    jdelony

882
1

cb9fc575e4 · common : use pimpl in debug.h to reduce header dependencies · Updated 2026-04-26 01:49:28 -05:00    jdelony

903
3