 leejetandmirek190
|
984eb0686c
|
perf(cuda): fuse MOSS modulation and direct RoPE F16 output (#2269)
Co-authored-by: mirek190 <mirek190@gmail.com>
|
2026-11-04 02:01:41 +08:00 |
|
|
Piotr Wilkin (ilintar)
|
6225bf8c88
|
perf: view qkv slices for moss instead of slice+reshape (#2232)
|
2026-11-03 23:46:35 +08:00 |
|
|
Mirek
|
f8ea2e162c
|
perf: use native SiTU for MOSS with composite fallback (#2210)
|
2026-11-03 23:41:09 +08:00 |
|
|
leejet
|
290148d2e3
|
feat: add custom modulate operator and ConvRot fusion (#2258)
|
2026-11-02 01:19:28 +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
|
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)
|
2026-10-30 00:38:14 +08:00 |
|
|
Daniele
|
8ee08e3cb9
|
perf: use fused ggml_geglu_swapped for GEGLU (#2245)
|
2026-10-29 23:18:10 +08:00 |
|
|
leejet
|
77b72a2efe
|
perf: accelerate INT8 diffusion inference with CUDA fusions (#2235)
|
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
|
be486498ea
|
fix: enable segmented execution for FLUX.3 (#2224)
|
2026-10-27 00:32:18 +08:00 |
|
|
leejet
|
e41fbcf455
|
feat(server): add recipe constraints and update generation workspace (#2217)
|
2026-10-26 01:01:43 +08:00 |
|
|
Mirekandleejet
|
e6370bf093
|
perf: avoid redundant MOSS QKV and value layout copies (#2209)
Co-authored-by: leejet <leejet714@gmail.com>
|
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 |
|
|
leejet
|
ea209dbfcf
|
perf: cache CPU position embeddings across DiT runners (#2211)
|
2026-10-25 05:18:30 +08:00 |
|
|
SolicTous
|
b29cad540a
|
feat: add control-net z-image (#2187)
|
2026-10-25 03:11:57 +08:00 |
|
|
losewayy
|
4c870cc60c
|
fix: balance singleton tails in short Wan vae decode plans (#2197)
|
2026-10-25 03:06:58 +08:00 |
|
|
Adrian Ciprian Iancu
|
50a64f075e
|
fix: skip metadata alpha for LoRAs with baked_scale (#2202)
|
2026-10-25 03:02:28 +08:00 |
|
|
leejet
|
e0ed724e28
|
chore: format code
|
2026-10-24 01:23:42 +08:00 |
|
|
losewayy
|
40b1788476
|
feat: allow grouped temporal decoding for Wan2.2 TI2V 5B (#2196)
|
2026-10-24 01:09:05 +08:00 |
|
|
leejet
|
cd7f986058
|
refactor: centralize model factories and defaults (#2198)
|
2026-10-24 01:03:05 +08:00 |
|
|
Philip Sorst
|
baf45b49e2
|
fix: forward ref_image_args to the video pipeline (#2177)
|
2026-10-24 00:57:24 +08:00 |
|
|
leejet
|
fd21ecfcee
|
refactor: use one placeholder token per image in LLM inputs (#2191)
|
2026-10-23 00:45:33 +08:00 |
|
|
leejet
|
2e2ba3faec
|
feat: add Fibo 2.0 Dev and Edit Turbo support (#2180)
|
2026-10-21 23:32:19 +08:00 |
|
|
leejet
|
ab8d5ea8c3
|
feat: support Diffusers-format MiniMax-H3 LoRAs (#2159)
|
2026-10-18 14:50:36 +08:00 |
|
|
leejet
|
68285df319
|
feat: add moss-image support (#2148)
|
2026-10-17 03:38:38 +08:00 |
|
|
leejet
|
af0fc72c96
|
chore: sync ggml and remove obsolete conv1d workarounds (#2147)
|
2026-10-17 02:39:51 +08:00 |
|
|
leejet
|
8cd731e7b4
|
feat: add FLUX.3 Image Klein and Klein Base support (#2143)
|
2026-10-16 02:42:07 +08:00 |
|
|
akleine
|
2891378bb1
|
fix: prevent core dump in CPU backend if using FP8 (e4m3) tensors (#2140)
|
2026-10-15 22:26:27 +08:00 |
|
|
Gabriele Lozupone
|
792bcd344c
|
feat: add MoSR upscaler support (#2128)
|
2026-10-12 22:38:38 +08:00 |
|
|
leejet
|
2988060a43
|
feat: add TAEQI2.1 support for Qwen Image 2.1 (#2123)
|
2026-10-11 00:01:18 +08:00 |
|
|
Daniele
|
1b0ba10893
|
feat: run conditional and unconditional CFG in one batched UNet forward (#2085)
|
2026-10-10 23:35:58 +08:00 |
|
|
leejet
|
f89d9b13d7
|
feat: add Iris-3B text-to-image support (#2117)
|
2026-10-10 01:39:49 +08:00 |
|
|
leejet
|
f2f168fa70
|
fix: preserve singleton output dimension when merging LoRA (#2116)
|
2026-10-10 01:12:03 +08:00 |
|
|
leejet
|
228c707fde
|
perf: use ggml rope apply op on supported backends (#2113)
|
2026-10-09 04:32:36 +08:00 |
|
|
losewayy
|
0963357214
|
fix: keep MiniMax-H3 VAE weights resident across temporal chunks (#2103)
|
2026-10-06 19:40:48 +08:00 |
|
|
Aleksei Pakhomov
|
bbdcad48dd
|
feat: add Z-Image L2P support (#2075)
|
2026-10-06 17:23:21 +08:00 |
|
|
losewayyandleejet
|
b6947b8bdc
|
fix: avoid f16 overflow in Z-Image quantized matmuls on CUDA (#2074)
Co-authored-by: leejet <leejet714@gmail.com>
|
2026-10-06 17:19:56 +08:00 |
|
|
Daniele
|
9db0db6570
|
perf: avoid redundant permute+cont in non-interleaved RoPE (#2102)
|
2026-10-06 16:54:55 +08:00 |
|
|
leejet
|
ede32a621e
|
fix: load INT8 convrot LLM embeddings correctly (#2067)
|
2026-09-27 19:46:21 +08:00 |
|
|
leejet
|
47e83d7136
|
sync: update ggml
|
2026-09-27 18:40:09 +08:00 |
|
|
leejet
|
f8890b9db3
|
feat: add Ming-Image Design support (#2063)
|
2026-09-27 17:24:13 +08:00 |
|
|
leejet
|
2f886889e6
|
refactor: align PixArt weights with upstream layout (#2061)
|
2026-09-26 02:03:05 +08:00 |
|
|
leejet
|
19bbbca1c7
|
refactor: define VAE tile dimensions in image pixels (#2059)
|
2026-09-25 18:20:32 +08:00 |
|
|
losewayy
|
39ada0863b
|
feat: add PixArt model family support (#2047)
|
2026-09-25 18:18:55 +08:00 |
|
|
leejet
|
510bccf330
|
fix: map Qwen Image 2.1 LoRAs to fused MLP weights (#2057)
|
2026-09-25 16:30:34 +08:00 |
|
|
leejet
|
0a9340c599
|
fix: scale Qwen Image 2.1 VAE convolutions (#2054)
|
2026-09-25 15:52:35 +08:00 |
|
|
leejet
|
740c7ae193
|
feat: add configurable Qwen cache types and early cache scheduling (#2045)
|
2026-09-25 01:51:01 +08:00 |
|
|
leejet
|
88411ef1e0
|
refactor: centralize circular RoPE and extend image model support (#2039)
|
2026-09-24 01:45:14 +08:00 |
|