update readme and some debug functions

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

README.md

@@ -1,8 +1,14 @@
 <div align="center">
-<img src="https://github.com/vladmandic/sdnext/raw/master/html/logo-transparent.png" width=200 alt="SD.Next">
+<img src="https://github.com/vladmandic/sdnext/raw/master/html/logo-transparent.png" width=200 alt="SD.Next: AI art generator logo">
 
-# SD.Next: All-in-one WebUI for AI generative image and video creation and captioning  
+# SD.Next: All-in-one WebUI
 
+SD.Next is a powerful, open-source WebUI app for AI image and video generation, built on Stable Diffusion and supporting dozens of advanced models. Create, caption, and process images and videos with a modern, cross-platform interface—perfect for artists, researchers, and AI enthusiasts.
+
+
+![Stars](https://img.shields.io/github/stars/vladmandic/sdnext?style=social)
+![Forks](https://img.shields.io/github/forks/vladmandic/sdnext?style=social)
+![Contributors](https://img.shields.io/github/contributors/vladmandic/sdnext)
 ![Last update](https://img.shields.io/github/last-commit/vladmandic/sdnext?svg=true)
 ![License](https://img.shields.io/github/license/vladmandic/sdnext?svg=true)
 [![Discord](https://img.shields.io/discord/1101998836328697867?logo=Discord&svg=true)](https://discord.gg/VjvR2tabEX)
@@ -17,61 +23,63 @@
 ## Table of contents
 
 - [Documentation](https://vladmandic.github.io/sdnext-docs/)
-- [SD.Next Features](#sdnext-features)
+- [SD.Features](#features--capabilities)
-- [Model support](#model-support)
+- [Supported AI Models](#supported-ai-models)
-- [Platform support](#platform-support)
+- [Supported Platforms & Hardware](#supported-platforms--hardware)
 - [Getting started](#getting-started)
 
-## SD.Next Features
+### Screenshot: Desktop interface
 
-All individual features are not listed here, instead check [ChangeLog](CHANGELOG.md) for full list of changes
+<div align="center">
-- Fully localized:
+<img src="https://github.com/vladmandic/sdnext/raw/dev/html/screenshot-robot.jpg" alt="SD.Next: AI art generator desktop interface screenshot" width="90%">
-  ▹ **English | Chinese | Russian | Spanish | German | French | Italian | Portuguese | Japanese | Korean**  
+</div>
-- Desktop and Mobile support!  
+
-- Multiple [diffusion models](https://vladmandic.github.io/sdnext-docs/Model-Support/)!  
+### Screenshot: Mobile interface
-- Multi-platform!  
+
- ▹ **Windows | Linux | MacOS | nVidia CUDA | AMD ROCm | Intel Arc / IPEX XPU | DirectML | OpenVINO | ONNX+Olive | ZLUDA**
+<div align="center">
+<img src="https://github.com/user-attachments/assets/ced9fe0c-d2c2-46d1-94a7-8f9f2307ce38" alt="SD.Next: AI art generator mobile interface screenshot" width="35%">
+</div>
+</div>
+
+<br>
+
+## Features & Capabilities
+
+SD.Next is feature-rich with a focus on performance, flexibility, and user experience. Key features include:
+- [Multi-platform](#platform-support!  
+- Many [diffusion models](https://vladmandic.github.io/sdnext-docs/Model-Support/)!  
+- Fully localized to ~15 languages and with support for many [UI themes](https://vladmandic.github.io/sdnext-docs/Themes/)!
+- [Desktop](#screenshot-desktop-interface) and [Mobile](#screenshot-mobile-interface) support!  
 - Platform specific auto-detection and tuning performed on install  
-- Optimized processing with latest `torch` developments with built-in support for model compile and quantize  
-  Compile backends: *Triton | StableFast | DeepCache | OneDiff | TeaCache | etc.*  
-  Quantization methods: *SDNQ | BitsAndBytes | Optimum-Quanto | TorchAO / LayerWise*  
-- **Captioning** with 150+ **OpenCLiP** models, **Tagger** with **WaifuDiffusion** and **DeepDanbooru** models, and 20+ built-in **VLMs**  
 - Built in installer with automatic updates and dependency management  
 
-<br>
+### Unique features
 
-**Desktop** interface  
+SD.Next includes many features not found in other WebUIs, such as:
-<div align="center">
+- **SDNQ**: State-of-the-Art quantization engine
-<img src="https://github.com/vladmandic/sdnext/raw/dev/html/screenshot-robot.jpg" alt="screenshot-modernui-desktop" width="90%">
+  Use pre-quantized or run with quantizaion on-the-fly for up to 4x VRAM reduction with no or minimal quality and performance impact  
-</div>
+- **Balanced Offload**: Dynamically balance CPU and GPU memory to run larger models on limited hardware
-
+- **Captioning** with 150+ **OpenCLiP** models, **Tagger** with **WaifuDiffusion** and **DeepDanbooru** models, and 25+ built-in **VLMs**  
-**Mobile** interface  
+- **Image Processing** with full image correction color-grading suite of tools  
-<div align="center">
-<img src="https://github.com/user-attachments/assets/ced9fe0c-d2c2-46d1-94a7-8f9f2307ce38" alt="screenshot-modernui-mobile" width="35%">
-</div>
-
-For screenshots and information on other available themes, see [Themes](https://vladmandic.github.io/sdnext-docs/Themes/)
 
 <br>
 
-## Model support
+## Supported AI Models
 
 SD.Next supports broad range of models: [supported models](https://vladmandic.github.io/sdnext-docs/Model-Support/) and [model specs](https://vladmandic.github.io/sdnext-docs/Models/)  
 
-## Platform support
+## Supported Platforms & Hardware
 
 - *nVidia* GPUs using **CUDA** libraries on both *Windows and Linux*  
-- *AMD* GPUs using **ROCm** libraries on *Linux*  
+- *AMD* GPUs using **ROCm** libraries on both *Linux and Windows*
-  Support will be extended to *Windows* once AMD releases ROCm for Windows  
+- *AMD* GPUs on Windows using **ZLUDA** libraries  
 - *Intel Arc* GPUs using **OneAPI** with *IPEX XPU* libraries on both *Windows and Linux*  
+- Any *CPU/GPU* or device compatible with **OpenVINO** libraries on both *Windows and Linux*  
 - Any GPU compatible with *DirectX* on *Windows* using **DirectML** libraries  
-  This includes support for AMD GPUs that are not supported by native ROCm libraries  
-- Any GPU or device compatible with **OpenVINO** libraries on both *Windows and Linux*  
 - *Apple M1/M2* on *OSX* using built-in support in Torch with **MPS** optimizations  
 - *ONNX/Olive*  
-- *AMD* GPUs on Windows using **ZLUDA** libraries  
 
-Plus Docker container recipes for: [CUDA, ROCm, Intel IPEX and OpenVINO](https://vladmandic.github.io/sdnext-docs/Docker/)
+Plus **Docker** container recipes for: [CUDA, ROCm, Intel IPEX and OpenVINO](https://vladmandic.github.io/sdnext-docs/Docker/)
 
 ## Getting started
 
@@ -84,21 +92,37 @@ Plus Docker container recipes for: [CUDA, ROCm, Intel IPEX and OpenVINO](https:/
 > And for platform specific information, check out  
 > [WSL](https://vladmandic.github.io/sdnext-docs/WSL/) | [Intel Arc](https://vladmandic.github.io/sdnext-docs/Intel-ARC/) | [DirectML](https://vladmandic.github.io/sdnext-docs/DirectML/) | [OpenVINO](https://vladmandic.github.io/sdnext-docs/OpenVINO/) | [ONNX & Olive](https://vladmandic.github.io/sdnext-docs/ONNX-Runtime/) | [ZLUDA](https://vladmandic.github.io/sdnext-docs/ZLUDA/) | [AMD ROCm](https://vladmandic.github.io/sdnext-docs/AMD-ROCm/) | [MacOS](https://vladmandic.github.io/sdnext-docs/MacOS-Python/) | [nVidia](https://vladmandic.github.io/sdnext-docs/nVidia/) | [Docker](https://vladmandic.github.io/sdnext-docs/Docker/)
 
+### Quick Start
+
+```shell
+git clone https://github.com/vladmandic/sdnext
+cd sdnext
+./webui.sh # Linux/Mac
+webui.bat  # Windows
+webui.ps1  # PowerShell
+```
+
 > [!WARNING]
 > If you run into issues, check out [troubleshooting](https://vladmandic.github.io/sdnext-docs/Troubleshooting/) and [debugging](https://vladmandic.github.io/sdnext-docs/Debug/) guides  
 
+
+## Community & Support
+
+If you're unsure how to use a feature, best place to start is [Docs](https://vladmandic.github.io/sdnext-docs/) and if its not there,  
+check [ChangeLog](https://vladmandic.github.io/sdnext-docs/CHANGELOG/) for when feature was first introduced as it will always have a short note on how to use it  
+
+And for any question, reach out on [Discord](https://discord.gg/VjvR2tabEX) or open an [issue](https://github.com/vladmandic/sdnext/issues) or [discussion](https://github.com/vladmandic/sdnext/discussions)  
+
 ### Contributing
 
 Please see [Contributing](CONTRIBUTING) for details on how to contribute to this project  
-And for any question, reach out on [Discord](https://discord.gg/VjvR2tabEX) or open an [issue](https://github.com/vladmandic/sdnext/issues) or [discussion](https://github.com/vladmandic/sdnext/discussions)  
 
-### Credits
+## License & Credits
 
+- SD.Next is licensed under the [Apache License 2.0](LICENSE.txt)
 - Main credit goes to [Automatic1111 WebUI](https://github.com/AUTOMATIC1111/stable-diffusion-webui) for the original codebase  
-- Additional credits are listed in [Credits](https://github.com/AUTOMATIC1111/stable-diffusion-webui/#credits)  
-- Licenses for modules are listed in [Licenses](html/licenses.html)  
 
-### Evolution
+## Evolution
 
 <a href="https://star-history.com/#vladmandic/sdnext&Date">
   <picture width=640>
@@ -109,9 +133,4 @@ And for any question, reach out on [Discord](https://discord.gg/VjvR2tabEX) or o
 
 - [OSS Stats](https://ossinsight.io/analyze/vladmandic/sdnext#overview)
 
-### Docs
-
-If you're unsure how to use a feature, best place to start is [Docs](https://vladmandic.github.io/sdnext-docs/) and if its not there,  
-check [ChangeLog](https://vladmandic.github.io/sdnext-docs/CHANGELOG/) for when feature was first introduced as it will always have a short note on how to use it  
-
 <br>

TODO.md

@@ -2,20 +2,22 @@
 
 ## Release
 
-- Update **README**
 - Bumb packages
 - Implement `unload_auxiliary_models`
 - Release **Launcher**
 - Release **Enso**
 - Update **ROCm**
 - Tips **Color Grading**
+- Tips **Latent Corrections**
 
 ## Internal
 
+- Integrate: [Depth3D](https://github.com/vladmandic/sd-extension-depth3d)
 - Feature: Color grading in processing
 - Feature: RIFE update
 - Feature: RIFE in processing
 - Feature: SeedVR2 in processing
+- Feature: Add video models to `Reference`
 - Deploy: Lite vs Expert mode
 - Engine: [mmgp](https://github.com/deepbeepmeep/mmgp)
 - Engine: `TensorRT` acceleration

launch.py

@@ -308,6 +308,7 @@ def main():
         log.warning('Restart is recommended due to packages updates...')
     t_server = time.time()
     t_monitor = time.time()
+
     while True:
         try:
             alive = uv.thread.is_alive()
@@ -326,8 +327,10 @@ def main():
         if float(monitor_rate) > 0 and t_current - t_monitor > float(monitor_rate):
             log.trace(f'Monitor: {get_memory_stats(detailed=True)}')
             t_monitor = t_current
-            from modules.api.validate import get_api_stats
+            # from modules.api.validate import get_api_stats
-            get_api_stats()
+            # get_api_stats()
+            # from modules import memstats
+            # memstats.get_objects()
         if not alive:
             if uv is not None and uv.wants_restart:
                 clean_server()

modules/cmd_args.py

@@ -74,8 +74,8 @@ def add_diag_args(p):
     p.add_argument('--safe', default=env_flag("SD_SAFE", False), action='store_true', help="Run in safe mode with no user extensions")
     p.add_argument('--test', default=env_flag("SD_TEST", False), action='store_true', help="Run test only and exit")
     p.add_argument('--version', default=False, action='store_true', help="Print version information")
-    p.add_argument("--monitor", default=os.environ.get("SD_MONITOR", -1), help="Run memory monitor, default: %(default)s")
+    p.add_argument("--monitor", type=float, default=float(os.environ.get("SD_MONITOR", -1)), help="Run memory monitor, default: %(default)s")
-    p.add_argument("--status", default=os.environ.get("SD_STATUS", -1), help="Run server is-alive status, default: %(default)s")
+    p.add_argument("--status", type=float, default=float(os.environ.get("SD_STATUS", -1)), help="Run server is-alive status, default: %(default)s")
 
 
 def add_log_args(p):

modules/memstats.py

@@ -1,9 +1,11 @@
 import re
 import sys
 import os
+import types
+from collections import deque
 import psutil
 import torch
-from modules import shared, errors
+from modules import shared, errors, devices
 from modules.logger import log
 
 
@@ -130,28 +132,53 @@ def reset_stats():
 class Object:
     pattern = r"'(.*?)'"
 
+    def get_size(self, obj, seen=None):
+        size = sys.getsizeof(obj)
+        if seen is None:
+            seen = set()
+        obj_id = id(obj)
+        if obj_id in seen:
+            return 0  # Avoid double counting
+        seen.add(obj_id)
+        if isinstance(obj, dict):
+            size += sum(self.get_size(k, seen) + self.get_size(v, seen) for k, v in obj.items())
+        elif isinstance(obj, (list, tuple, set, frozenset, deque)):
+            size += sum(self.get_size(i, seen) for i in obj)
+        return size
+
     def __init__(self, name, obj):
         self.id = id(obj)
         self.name = name
         self.fn = sys._getframe(2).f_code.co_name
-        self.size = sys.getsizeof(obj)
         self.refcount = sys.getrefcount(obj)
         if torch.is_tensor(obj):
             self.type = obj.dtype
             self.size = obj.element_size() * obj.nelement()
         else:
             self.type = re.findall(self.pattern, str(type(obj)))[0]
-            self.size = sys.getsizeof(obj)
+            self.size = self.get_size(obj)
     def __str__(self):
         return f'{self.fn}.{self.name} type={self.type} size={self.size} ref={self.refcount}'
 
 
-def get_objects(gcl=None, threshold:int=0):
+def get_objects(gcl=None, threshold:int=1024*1024):
+    devices.torch_gc(force=True)
     if gcl is None:
+        # gcl = globals()
         gcl = {}
+        log.trace(f'Memory: modules={len(sys.modules)}')
+        for _module_name, module in sys.modules.items():
+            try:
+                if not isinstance(module, types.ModuleType):
+                    continue
+                namespace = vars(module)
+                gcl.update(namespace)
+            except Exception:
+                pass # Some modules may not allow introspection
     objects = []
     seen = []
 
+    log.trace(f'Memory: items={len(gcl)} threshold={threshold}')
     for name, obj in gcl.items():
         if id(obj) in seen:
             continue
@@ -169,6 +196,6 @@ def get_objects(gcl=None, threshold:int=0):
 
     objects = sorted(objects, key=lambda x: x.size, reverse=True)
     for obj in objects:
-        log.trace(obj)
+        log.trace(f'Memory: {obj}')
 
     return objects