83be0d77d7
zluda hip sdk 6.4
2025-08-15 20:24:50 +09:00
1ce9aae143
lint
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-14 16:57:20 -04:00
001716633d
fix namegen
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-14 16:40:42 -04:00
2318f97991
manual set dtype via api
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-14 15:35:50 -04:00
6e33723639
update models page
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-14 13:40:32 -04:00
1a59205cc1
add qwen-image taesd preview
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-14 11:23:25 -04:00
ad5881b9ce
add network delete
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-14 10:32:00 -04:00
b74e851fae
update changelog
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-14 10:15:26 -04:00
4ef1e56622
simplify namegen seq
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-14 10:14:16 -04:00
e3bc172e0a
update requirements
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-14 09:51:36 -04:00
1a60116dce
fix zavychromaxl
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-14 09:45:49 -04:00
e4cc2b1ee0
fix gallery mtime
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-14 09:42:58 -04:00
068f63badc
add model parallel load option
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 21:18:57 -04:00
706bc9e2e7
cleanup
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 20:55:26 -04:00
ff2c03538d
add mtime to reference models
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 20:53:39 -04:00
5cd88e93ed
Merge pull request #4115 from CalamitousFelicitousness/patch-1
...
LoRA load Torch tuple and string version checking
2025-08-13 20:25:43 -04:00
114d097bb4
add wildcards to networks ui
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 20:23:35 -04:00
70672d3267
Move tuple/string versioning fix for BnB to loader.py
2025-08-14 00:31:10 +01:00
afe900537d
Revert lora_load.py commit
2025-08-14 00:25:56 +01:00
6123655ac2
LoRA load Torch tuple and string version checking
...
Due to BitsandBytes trying to use tuple and comparison to check Torch version which is given as a string, using LoRA with a quantized model results in a TypeError. This commit adds support for both.
2025-08-13 23:39:18 +01:00
e30e8cbf4f
fix loading custom transformer safetensors
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 16:27:52 -04:00
41ae06bd90
fix loading custom t5
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 16:06:25 -04:00
8ca74d0cd2
SDNQ rename unused param_name arg to op
2025-08-13 22:10:30 +03:00
4d95cf47ce
restore default sampler on mismatch
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 14:49:11 -04:00
867c5abc5c
add qwen img2img and inpaint
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 14:35:06 -04:00
e8002df534
generate api endpoints allow set model
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 13:49:23 -04:00
0ae24f9916
fix ui checkbox/radio styling for non-default themes
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 13:12:20 -04:00
562799314d
change default hfcache folder
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 13:04:14 -04:00
8e9244939b
calc folder size
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 12:23:29 -04:00
532e89b4fa
add hunyuandit-distilled
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-13 10:56:13 -04:00
cb0c5414a3
SDNQ use uint with minimum_bits <= 4
2025-08-13 00:37:44 +03:00
7085db9add
Update changelog
2025-08-13 00:17:15 +03:00
15cb8fe9f8
SDNQ add modules_dtype_dict and fix Qwen Image with quants less than 5 bits
2025-08-13 00:07:36 +03:00
a52316b24c
fix flex.2
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-12 15:55:43 -04:00
ffc317c68c
update changelog
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-12 15:09:08 -04:00
33b92d9ad9
improve prompt enhance system prompt
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-12 15:09:08 -04:00
863e172aad
add Qwen/Qwen2.5-VL-3B-Instruct
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-12 15:09:08 -04:00
2aa917b58e
add /sdapi/v1/modules endpoint
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-12 15:09:08 -04:00
362ec0d412
Fix Chroma quantization
2025-08-12 22:07:52 +03:00
a36676a5a9
cleanup lora api
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-12 08:49:38 -04:00
669f9874a0
increase ui default timeouts
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-12 08:19:22 -04:00
28ad5bb0c3
image download use real filename
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-12 08:15:29 -04:00
fe38da3086
cleanup
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-11 17:04:42 -04:00
4317f23062
update modernui
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-11 16:59:40 -04:00
8b9ba82982
Use generic model loader with lumina2
2025-08-11 23:57:40 +03:00
9daa30f371
Add torch_dtype to transformer.from_single_file
2025-08-11 23:37:07 +03:00
6312b3d0ac
add Qwen3-4B-Instruct-2507 llm and Flash-FlowMatch scheduler
...
Signed-off-by: Vladimir Mandic <mandic00@live.com >
2025-08-11 16:35:16 -04:00
232a631be5
Use pre-mode quant on custom transformer files
2025-08-11 23:31:06 +03:00
9992338187
sdnq fix convs
2025-08-11 23:24:13 +03:00
26461f1d8d
fix conv in8 matmul
2025-08-11 23:15:30 +03:00
Seunghoon Lee
Vladimir Mandic
CalamitousFelicitousness
Disty0