Default Branch

1bfd842b84 · feat(gallery): add vllm-cpp entries for laya, cua-s1-forms, and gliner2.5 (#12240) · Updated 2026-09-25 06:20:52 -04:00

Branches

504751c493 · fix(oci): silence gosec on the validated link-copy file ops · Updated 2026-07-17 14:50:15 -04:00

1025
4

3159ed0637 · docs(paged): record P6 fp8-KV measured NO-GO - throughput dead end, capacity-play open · Updated 2026-07-02 19:01:40 -04:00

1250
321

6815bb2034 · chore(recon): re-pin to squashed release commits (voice 1db1759, face e22260d) · Updated 2026-06-29 11:22:32 -04:00

1351
29

ca4222e3e3 · chore: bump RFDETR_VERSION to ecf64d7 (current rf-detr.cpp main) · Updated 2026-05-27 18:32:43 -04:00

1635
2

20854bc000 · fix(llama-cpp): track upstream rename checkpoint_every_nt -> checkpoint_min_step · Updated 2026-05-25 17:46:54 -04:00

1665
2

11c5fd677d · fix(gallery/ltx-2.3): add vae_decode_only:false for i2v / flf2v · Updated 2026-05-25 15:33:55 -04:00

1671
2

8fbf18490e · fix: remove deprecated cosign bundle flag from backend merge workflow · Updated 2026-05-22 18:16:44 -04:00

1694
2

b6fed26271 · chore(turboquant): retreat pin to 4c1c3ac0 to skip fork GPU regression · Updated 2026-05-21 11:54:38 -04:00

1712
2

9d42a16c20 · ci: publish base images to ci-cache instead of localai-base · Updated 2026-05-06 11:13:06 -04:00

1872
3

50580a84ae · fix(ci): switch apt mirror per runner — azure on github-hosted, kernel.org on self-hosted · Updated 2026-05-03 18:59:26 -04:00

1905
0
Included

3280b9a287 · fix(distributed): per-replica backend logs (store aggregation + UI) · Updated 2026-04-27 16:55:24 -04:00

1942
0
Included

d9d7b5c29b · docs(readme): add April 2026 highlights to Latest News · Updated 2026-04-23 16:47:06 -04:00

1995
0
Included

5f7a0c3b26 · chore(turboquant): bump fork pin to rebase/upstream-sync-april-2026 · Updated 2026-04-22 16:01:49 -04:00

2009
1

798b5b2d84 · chore(turboquant): bump fork to 4d24ad87 and patch ggml-hip for new f16-turbo fattn-vec instances · Updated 2026-04-22 03:13:47 -04:00

2013
1

b27de08fff · chore(gallery): fixup wan · Updated 2026-04-19 17:31:22 -04:00

2052
0
Included

44e7d9806b · fix(distributed): stop queue loops on agent nodes + dead-letter cap · Updated 2026-04-19 17:27:05 -04:00

2057
8

fbc93b0a34 · fix(llama-cpp): default rms_norm_eps for Gemma 3 GGUFs missing the key · Updated 2026-04-19 12:15:26 -04:00

2055
1

cd56a05c3e · ci(vllm): disable tests-vllm-grpc job (heterogeneous runners) · Updated 2026-04-13 03:46:57 -04:00

2122
16

5fe87cb0d5 · feat: upgrade banner with Upgrade All button, detect pre-existing backends · Updated 2026-04-11 18:11:23 -04:00

2131
8

6e11f882f7 · feat(turboquant.cpp): add new backend · Updated 2026-04-03 16:57:15 -04:00

2196
1