add flex.1-alpha

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

CHANGELOG.md

@@ -13,13 +13,20 @@
     - [localization](https://github.com/vladmandic/sdnext/wiki/Locale) documentation  
   - **UI**:  
     - force browser cache-invalidate on page load  
+- **Models**
+  - [Ostris Flex.1-Alpha](https://huggingface.co/ostris/Flex.1-alpha)  
+    originally based on Flux.1-Schnell, but retrained and with different architecture  
+    result is model smaller than Flux.1-Dev, but with similar capabilities  
 - **Docs**
   - New [Outpaint](https://github.com/vladmandic/sdnext/wiki/Outpaint) step-by-step guide  
   - Updated [Docker](https://github.com/vladmandic/sdnext/wiki/Docker) guide  
     includes build and publish and both local and cloud examples  
+- **Docker** 
+  - updated **CUDA** receipe to `torch==2.6.0` with `cuda==12.6`  
+  - added **ROCm** receipe
+  - added **IPEX** receipe
+  - added **OpenVINO** receipe
 - **Backend**  
-  - **Docker** 
-    - updated CUDA image to `torch==2.6.0` with `cuda==12.6`  
   - **Torch**  
     - for **zluda** set default to `torch==2.6.0+cu118`  
     - for **openvino** set default to `torch==2.6.0+cpu`  

README.md

@@ -68,6 +68,8 @@ SD.Next supports broad range of models: [supported models](https://vladmandic.gi
 - *ONNX/Olive*  
 - *AMD* GPUs on Windows using **ZLUDA** libraries  
 
+Plus Docker container receipes for: [CUDA, ROCm, Intel IPEX and OpenVINO](https://vladmandic.github.io/sdnext-docs/Docker/)
+
 ## Getting started
 
 - Get started with **SD.Next** by following the [installation instructions](https://vladmandic.github.io/sdnext-docs/Installation/)  

html/reference.json

@@ -179,6 +179,13 @@
     "skip": true,
     "extras": "sampler: Default, cfg_scale: 3.5"
   },
+  "Ostris Flex.1 Alpha": {
+    "path": "ostris/Flex.1-alpha",
+    "preview": "ostris--Flex.1-alpha.jpg",
+    "desc": "Flex.1 alpha is a pre-trained base 8 billion parameter rectified flow transformer capable of generating images from text descriptions. It has a similar architecture to FLUX.1-dev, but with fewer double transformer blocks (8 vs 19)",
+    "skip": true,
+    "extras": "sampler: Default, cfg_scale: 3.5"
+  },
 
   "NVLabs Sana 1.6B 4k": {
     "path": "Efficient-Large-Model/Sana_1600M_4Kpx_BF16_diffusers",

modules/sd_detect.py

@@ -86,7 +86,7 @@ def detect_pipeline(f: str, op: str = 'model', warning=True, quiet=False):
                 pipeline = 'custom'
             if 'sd3' in f.lower():
                 guess = 'Stable Diffusion 3'
-            if 'flux' in f.lower():
+            if 'flux' in f.lower() or 'flex.1' in f.lower():
                 guess = 'FLUX'
                 if size > 11000 and size < 16000:
                     warn(f'Model detected as FLUX UNET model, but attempting to load a base model: {op}={f} size={size} MB')

scripts/mixture_of_diffusers.py

@@ -0,0 +1,37 @@
+import gradio as gr
+from modules import scripts, processing, shared
+
+
+class Script(scripts.Script):
+    def __init__(self):
+        super().__init__()
+        self.orig_pipe = None
+
+    def title(self):
+        return 'Mixture-of-Diffusers'
+
+    def show(self, is_img2img):
+        return shared.native
+
+    def ui(self, _is_img2img): # ui elements
+        with gr.Row():
+            gr.HTML('<a href="https://arxiv.org/abs/2302.02412">&nbsp Mixture-of-Diffusers</a><br>')
+        return []
+
+    def run(self, p: processing.StableDiffusionProcessing): # pylint: disable=arguments-differ, unused-argument
+        supported_model_list = ['sdxl']
+        if shared.sd_model_type not in supported_model_list:
+            shared.log.warning(f'MoD: class={shared.sd_model.__class__.__name__} model={shared.sd_model_type} required={supported_model_list}')
+            return None
+        self.orig_pipe = shared.sd_model
+
+        shared.log.info(f'MoD: ')
+
+
+    def after(self, p: processing.StableDiffusionProcessing, processed: processing.Processed): # pylint: disable=arguments-differ, unused-argument
+        if self.orig_pipe is None:
+            return processed
+        if shared.sd_model_type == "sdxl":
+            shared.sd_model = self.orig_pipe
+        self.orig_pipe = None
+        return processed