leejet
|
3be23788fc
|
ci: use ccache to fix Windows CUDA command parsing failures
master-1043-3be2378
|
2026-11-03 00:20:46 +08:00 |
|
leejet
|
b4d6b83af6
|
fix: support integer LoRA scalar conversion to FP32 (#2263)
|
2026-11-03 00:07:40 +08:00 |
|
cumal
|
250ea6e385
|
fix: layer split across cuda (#2262)
|
2026-11-03 00:03:43 +08:00 |
|
leejet
|
1350cf5097
|
ci: cache CUDA compilation for Windows and Docker builds
|
2026-11-02 23:51:33 +08:00 |
|
leejet
|
290148d2e3
|
feat: add custom modulate operator and ConvRot fusion (#2258)
master-1039-290148d
|
2026-11-02 01:19:28 +08:00 |
|
leejet
|
133087deb8
|
fix: prefer discrete GPUs in auto-fit device selection (#2259)
master-1038-133087d
|
2026-11-02 01:09:36 +08:00 |
|
leejet
|
5d33b1bd2b
|
ci: trigger builds on CUDA header changes
master-1037-5d33b1b
|
2026-11-02 00:23:44 +08:00 |
|
leejet
|
62c54439c8
|
fix: include missing cassert headers in SageAttention CUDA code
|
2026-11-02 00:17:49 +08:00 |
|
leejet
|
fe2682ba16
|
fix: retry Vulkan capacity checks after memory reclamation (#2256)
|
2026-11-01 22:55:56 +08:00 |
|
leejet
|
df0fa26a61
|
fix: update CUTLASS license paths in Docker and Windows CUDA packaging
|
2026-11-01 22:22:04 +08:00 |
|
leejet
|
5356982584
|
fix: chunk MiniMax-H3 MLP to respect backend buffer limits (#2257)
|
2026-11-01 22:08:29 +08:00 |
|
leejet
|
45a63ac460
|
refactor: move model-specific operators and fusions into backend extensions (#2253)
|
2026-11-01 21:18:37 +08:00 |
|
leejet
|
c6dfa9def1
|
perf(vulkan): update ggml for H3 inference optimizations (#2252)
master-1031-c6dfa9d
|
2026-11-01 00:26:59 +08:00 |
|
leejet
|
0a067120b3
|
perf: optimize Vulkan kernels, graph fusion, and workspace reuse (#2251)
master-1030-0a06712
|
2026-10-31 14:55:58 +08:00 |
|
leejet
|
e239e4b0fd
|
refactor: replace manual gated activations with native ggml GLU (#2250)
|
2026-10-30 01:32:46 +08:00 |
|
leejet
|
37bc88b5e9
|
perf: optimize H3 inference and segmented weight prefetch (#2240)
master-1028-37bc88b
|
2026-10-30 00:38:14 +08:00 |
|
dubs-dev
|
e7b7ec6d6f
|
fix: detect SDXL when the UNet is loaded standalone from diffusers tensor names (#2247)
master-1027-e7b7ec6
|
2026-10-29 23:33:44 +08:00 |
|
vmobilis
|
997478a4cb
|
refactor: make_hires_sigma_schedule() simplification (#2244)
master-1026-997478a
|
2026-10-29 23:21:20 +08:00 |
|
Daniele
|
8ee08e3cb9
|
perf: use fused ggml_geglu_swapped for GEGLU (#2245)
master-1025-8ee08e3
|
2026-10-29 23:18:10 +08:00 |
|
leejet
|
77b72a2efe
|
perf: accelerate INT8 diffusion inference with CUDA fusions (#2235)
master-1024-77b72a2
|
2026-10-28 22:02:46 +08:00 |
|
Wagner Bruna
|
8cbe905fdd
|
refactor: use a distinct version id for Chroma (#2237)
|
2026-10-28 22:02:18 +08:00 |
|
leejet
|
687915518e
|
ci: complete Node 24 migration for workflows and Docker builds
|
2026-10-28 01:53:21 +08:00 |
|
project516
|
d09cb4c63a
|
ci: update workflow actions for Node 24 (#2230)
|
2026-10-28 01:42:45 +08:00 |
|
Crystal Sight
|
209b28c3cd
|
feat(server): add recipe model defaults and idempotent model loading (#2221)
|
2026-10-28 01:36:23 +08:00 |
|
project516
|
9bfc4d1089
|
fix: honor the requested hires sampling step count (#2227)
|
2026-10-28 01:33:03 +08:00 |
|
project516
|
2032526e8c
|
fix: include integrated GPUs in automatic memory planning (#2228)
|
2026-10-28 01:28:11 +08:00 |
|
leejet
|
be486498ea
|
fix: enable segmented execution for FLUX.3 (#2224)
master-1017-be48649
|
2026-10-27 00:32:18 +08:00 |
|
leejet
|
bb6b128b2d
|
fix: avoid repeated seeks when skipping small GGUF metadata values (#2223)
master-1016-bb6b128
|
2026-10-27 00:24:43 +08:00 |
|
leejet
|
e41fbcf455
|
feat(server): add recipe constraints and update generation workspace (#2217)
master-1015-e41fbcf
|
2026-10-26 01:01:43 +08:00 |
|
leejet
|
012c7f8866
|
ci: trigger builds on recipe JSON changes
master-1014-012c7f8
|
2026-10-25 19:39:32 +08:00 |
|
leejet
|
a5e48c66aa
|
chore: reduce large recipe defaults to 1024x1024 except PiD
|
2026-10-25 19:37:41 +08:00 |
|
leejet
|
b29962db7a
|
sync: update frontend
|
2026-10-25 19:30:23 +08:00 |
|
Crystal Sight
|
0f76bf5c1e
|
feat(server): accept ref_videos, ref_audios, ref_video_audios in sdcpp vid_gen (#2194)
|
2026-10-25 17:18:47 +08:00 |
|
leejet
|
bb6a0ccfa3
|
feat: support FLUX.3 Diffusers LoRA tensor names (#2214)
master-1010-bb6a0cc
|
2026-10-25 16:55:45 +08:00 |
|
 Mirekandleejet
|
e6370bf093
|
perf: avoid redundant MOSS QKV and value layout copies (#2209)
Co-authored-by: leejet <leejet714@gmail.com>
master-1009-e6370bf
|
2026-10-25 16:54:51 +08:00 |
|
Mirek
|
2a5f81095d
|
perf: reuse shared modulation slices for moss image (#2208)
|
2026-10-25 16:00:26 +08:00 |
|
Wagner Bruna
|
d234ac62fa
|
feat: print sigma values before inference (#2205)
master-1007-d234ac6
|
2026-10-25 15:58:35 +08:00 |
|
Wagner Bruna
|
0886f11b6b
|
fix: skip reference image VAE encoding when latents have no consumer (#2207)
master-1006-0886f11
|
2026-10-25 05:19:32 +08:00 |
|
leejet
|
ea209dbfcf
|
perf: cache CPU position embeddings across DiT runners (#2211)
master-1005-ea209db
|
2026-10-25 05:18:30 +08:00 |
|
leejet
|
a85db96aa5
|
feat: return generation metadata with core results (#2206)
master-1004-a85db96
|
2026-10-25 03:18:03 +08:00 |
|
SolicTous
|
b29cad540a
|
feat: add control-net z-image (#2187)
master-1003-b29cad5
|
2026-10-25 03:11:57 +08:00 |
|
losewayy
|
4c870cc60c
|
fix: balance singleton tails in short Wan vae decode plans (#2197)
master-1002-4c870cc
|
2026-10-25 03:06:58 +08:00 |
|
Adrian Ciprian Iancu
|
50a64f075e
|
fix: skip metadata alpha for LoRAs with baked_scale (#2202)
master-1001-50a64f0
|
2026-10-25 03:02:28 +08:00 |
|
leejet
|
95ff762308
|
fix: update ggml to fix SageAttention builds on older GPUs (#2199)
master-1000-95ff762
|
2026-10-24 01:30:55 +08:00 |
|
leejet
|
e0ed724e28
|
chore: format code
master-999-e0ed724
|
2026-10-24 01:23:42 +08:00 |
|
losewayy
|
40b1788476
|
feat: allow grouped temporal decoding for Wan2.2 TI2V 5B (#2196)
master-998-40b1788
|
2026-10-24 01:09:05 +08:00 |
|
leejet
|
cd7f986058
|
refactor: centralize model factories and defaults (#2198)
master-997-cd7f986
|
2026-10-24 01:03:05 +08:00 |
|
Philip Sorst
|
baf45b49e2
|
fix: forward ref_image_args to the video pipeline (#2177)
master-996-baf45b4
|
2026-10-24 00:57:24 +08:00 |
|
leejet
|
fd21ecfcee
|
refactor: use one placeholder token per image in LLM inputs (#2191)
master-995-fd21ecf
|
2026-10-23 00:45:33 +08:00 |
|
leejet
|
e779d4f149
|
feat(server): support cancellation of running generation jobs (#2190)
master-994-e779d4f
|
2026-10-23 00:24:56 +08:00 |
|