fix lora with hidiffusion blocks

Signed-off-by: Vladimir Mandic <mandic00@live.com>

CHANGELOG.md

@@ -6,7 +6,10 @@
 
 Feature highlights include:  
 - **ModernUI** has quite some redesign which should make it more user friendly and easier to navigate plus several new UI themes!  
-- New models [WanAI Wan 2.1](https://wan.video/) for text-to-image workflows, [FreePix F-Lite](https://huggingface.co/Freepik/F-Lite), [Bria 3.2](https://huggingface.co/briaai/BRIA-3.2), [bigASP 2.5](https://civitai.com/models/1789765?modelVersionId=2025412)  
+  If you're still using StandardUI, give [ModernUI](https://vladmandic.github.io/sdnext-docs/Themes/) a try!  
+- New models:  
+  - [WanAI 2.2](https://wan.video/) in 5B and A14B variants for both *text-to-video* and *image-to-video* workflows as well as *text-to-image* workflow!  
+  - [FreePix F-Lite](https://huggingface.co/Freepik/F-Lite), [Bria 3.2](https://huggingface.co/briaai/BRIA-3.2), [bigASP 2.5](https://civitai.com/models/1789765?modelVersionId=2025412)  
 - Redesigned [LTXVideo](https://vladmandic.github.io/sdnext-docs/Video) interface with support for general video models plus optimized [FramePack](https://vladmandic.github.io/sdnext-docs/FramePack) and [LTXVideo](https://vladmandic.github.io/sdnext-docs/LTX) support  
 - Fully integrated nudity detection and optional censorship with [NudeNet](https://vladmandic.github.io/sdnext-docs/NudeNet)  
 - New background replacement and relightning methods using **Latent Bridge Matching** and new **PixelArt** processing filter  
@@ -35,8 +38,13 @@ For details, see [ChangeLog](https://github.com/vladmandic/automatic/blob/master
   - SD.Next [license](https://github.com/vladmandic/sdnext/blob/dev/LICENSE.txt) switched from **aGPL-v3.0** to **Apache-v2.0**  
     this means that SD.Next is now fully compatible with commercial and non-commercial use and redistribution regardless of modifications!  
 - **Models**
-  - [WanAI Wan 2.1](https://wan.video/) 1.3B and 14B T2V models for T2I workflows  
-    Wan 2.1 is originally designed for *video* workflows, Wan 2.1 model can now also be used for *text-to-image* workflows!  
+  - [WanAI Wan 2.2](https://github.com/Wan-Video/Wan2.2) both 5B and A14B variants, for both T2V and I2V support  
+    go to: *video -> generic -> wan -> pick variant*  
+    optimized support with *VACE*, etc. will follow soon  
+    *note*: quantization and offloading are highly recommended!  
+  - [WanAI Wan](https://wan.video/) T2V models for T2I workflows  
+    Wan is originally designed for *video* workflows, but now also be used for *text-to-image* workflows!  
+    Supports Wan2.1 in 1.3B and 14B variants and Wan2.2 in 5B and A14B variants  
     supports all standard features such as quantization, offloading, TAESD preview generation, LoRA support etc.  
     can also load unet/transformer fine-tunes in safetensors format using UNET loader  
     simply select in *networks -> models -> reference*  

html/reference.json

@@ -203,6 +203,20 @@
     "skip": true,
     "extras": "sampler: Default"
   },
+  "Wan-AI Wan2.2 5B": {
+    "path": "Wan-AI/Wan2.2-TI2V-5B-Diffusers",
+    "preview": "Wan-AI--Wan2.2.jpg",
+    "desc": "Wan2.2, offering more powerful capabilities, better performance, and superior visual quality. With Wan2.2, we have focused on incorporating the following technical innovations: MoE Architecture, Data Scalling, Cinematic Aesthetics, Efficient High-Definition Hybrid",
+    "skip": true,
+    "extras": "sampler: Default"
+  },
+  "Wan-AI Wan2.2 A14B": {
+    "path": "Wan-AI/Wan2.2-T2V-A14B-Diffusers",
+    "preview": "Wan-AI--Wan2.2.jpg",
+    "desc": "Wan2.2, offering more powerful capabilities, better performance, and superior visual quality. With Wan2.2, we have focused on incorporating the following technical innovations: MoE Architecture, Data Scalling, Cinematic Aesthetics, Efficient High-Definition Hybrid",
+    "skip": true,
+    "extras": "sampler: Default"
+  },
 
   "Freepik F-Lite": {
     "path": "Freepik/F-Lite",
@@ -211,6 +225,20 @@
     "skip": true,
     "extras": "sampler: Default, cfg_scale: 3.5"
   },
+  "Freepik F-Lite Texture": {
+    "path": "Freepik/F-Lite-Texture",
+    "preview": "Freepik--F-Lite.jpg",
+    "desc": "F Lite is a 10B parameter diffusion model created by Freepik and Fal, trained exclusively on copyright-safe and SFW content. The model was trained on Freepik's internal dataset comprising approximately 80 million copyright-safe images, making it the first publicly available model of this scale trained exclusively on legally compliant and SFW content.",
+    "skip": true,
+    "extras": "sampler: Default, cfg_scale: 3.5"
+  },
+  "Freepik F-Lite 7B": {
+    "path": "Freepik/F-Lite-7B",
+    "preview": "Freepik--F-Lite.jpg",
+    "desc": "F Lite is a 10B parameter diffusion model created by Freepik and Fal, trained exclusively on copyright-safe and SFW content. The model was trained on Freepik's internal dataset comprising approximately 80 million copyright-safe images, making it the first publicly available model of this scale trained exclusively on legally compliant and SFW content.",
+    "skip": true,
+    "extras": "sampler: Default, cfg_scale: 3.5"
+  },
 
   "NVLabs Sana 1.5 1.6B 1k": {
     "path": "Efficient-Large-Model/SANA1.5_1.6B_1024px_diffusers",

modules/lora/network_lora.py

@@ -3,6 +3,7 @@ import diffusers.models.lora as diffusers_lora
 import modules.lora.lyco_helpers as lyco_helpers
 import modules.lora.network as network
 from modules import devices
+from modules.errors import log
 
 
 class ModuleTypeLora(network.ModuleType):
@@ -26,8 +27,9 @@ class NetworkModuleLora(network.NetworkModule):
         if weight is None and none_ok:
             return None
         linear_modules = [torch.nn.Linear, torch.nn.modules.linear.NonDynamicallyQuantizableLinear, torch.nn.MultiheadAttention, diffusers_lora.LoRACompatibleLinear]
-        is_linear = type(self.sd_module) in linear_modules or self.sd_module.__class__.__name__ in {"NNCFLinear", "QLinear", "Linear4bit"}
-        is_conv = type(self.sd_module) in [torch.nn.Conv2d, diffusers_lora.LoRACompatibleConv] or self.sd_module.__class__.__name__ in {"NNCFConv2d", "QConv2d"}
+        typ = type(self.sd_module)
+        is_linear = typ in linear_modules or self.sd_module.__class__.__name__ in ["NNCFLinear", "QLinear", "Linear4bit"]
+        is_conv = (typ in [torch.nn.Conv2d, diffusers_lora.LoRACompatibleConv]) or (self.sd_module.__class__.__name__ in ["NNCFConv2d", "QConv2d"]) or (typ.__name__ in ['downsampler_block', 'upsampler_block'])
         if is_linear:
             weight = weight.reshape(weight.shape[0], -1)
             module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False)
@@ -43,7 +45,7 @@ class NetworkModuleLora(network.NetworkModule):
         elif is_conv and (key == "lora_up.weight" or key == "dyn_down"):
             module = torch.nn.Conv2d(weight.shape[1], weight.shape[0], (1, 1), bias=False)
         else:
-            raise AssertionError(f'Lora unsupported: linear={is_linear} conv={is_conv} key={key} layer={self.network_key} type={type(self.sd_module).__name__}')
+            raise AssertionError(f'Lora unsupported: key={key} layer={self.network_key} type={typ.__name__}')
         with torch.no_grad():
             if weight.shape != module.weight.shape:
                 weight = weight.reshape(module.weight.shape)

modules/video_models/models_def.py

@@ -140,6 +140,30 @@ models = {
     ],
     'WAN Video': [
         Model(name='None'),
+        Model(name='WAN 2.2 5B T2V',
+              url='https://huggingface.co/Wan-AI/Wan2.2-TI2V-5B-Diffusers',
+              repo='Wan-AI/Wan2.2-TI2V-5B-Diffusers',
+              repo_cls=diffusers.WanPipeline,
+              te_cls=transformers.T5EncoderModel,
+              dit_cls=diffusers.WanTransformer3DModel),
+        Model(name='WAN 2.2 5B I2V',
+              url='https://huggingface.co/Wan-AI/Wan2.2-TI2V-5B-Diffusers',
+              repo='Wan-AI/Wan2.2-TI2V-5B-Diffusers',
+              repo_cls=diffusers.WanPipeline,
+              te_cls=transformers.T5EncoderModel,
+              dit_cls=diffusers.WanTransformer3DModel),
+        Model(name='WAN 2.2 A14B T2V',
+              url='https://huggingface.co/Wan-AI/Wan2.2-T2V-A14B-Diffusers',
+              repo='Wan-AI/Wan2.2-T2V-A14B-Diffusers',
+              repo_cls=diffusers.WanPipeline,
+              te_cls=transformers.T5EncoderModel,
+              dit_cls=diffusers.WanTransformer3DModel),
+        Model(name='WAN 2.2 A14B I2V',
+              url='https://huggingface.co/Wan-AI/Wan2.2-I2V-A14B-Diffusers',
+              repo='Wan-AI/Wan2.2-T2V-A14B-Diffusers',
+              repo_cls=diffusers.WanPipeline,
+              te_cls=transformers.T5EncoderModel,
+              dit_cls=diffusers.WanTransformer3DModel),
         Model(name='WAN 2.1 1.3B T2V',
               url='https://huggingface.co/Wan-AI/Wan2.1-T2V-1.3B-Diffusers',
               repo='Wan-AI/Wan2.1-T2V-1.3B-Diffusers',