Skip to content

Commit 546132d

Browse files
bojrodevbojrodev
authored
Update README.md
1 parent 7144cab commit 546132d

1 file changed

Lines changed: 44 additions & 73 deletions

File tree

‎README.md‎

Lines changed: 44 additions & 73 deletions
Original file line numberDiff line numberDiff line change
@@ -1,15 +1,10 @@
1-
# Resolver: Native Android Client for WebUI Forge & ComfyUi (Flux, SDXL & Qwen)
1+
# Resolver: Native Android Client for WebUI Forge & ComfyUI
22

3-
![Version](https://img.shields.io/badge/Version-1.3-purple.svg)
43
![Platform](https://img.shields.io/badge/Platform-Android%2010+-green.svg)
54
![Backend](https://img.shields.io/badge/Backend-Forge%20Neo-blue)
65
![License](https://img.shields.io/badge/License-GPLv3-red.svg)
76

8-
**Resolver** is a high-performance, native Android interface for [Stable Diffusion WebUI Forge Neo](https://github.com/Haoming02/sd-webui-forge-classic/tree/neo).
9-
10-
**New in v1.3:** Resolver now features a dedicated **ComfyUI Engine** for workflow execution, centralized **Configuration (CFG)** for seamless Local/External switching, a remote "Kill Switch", **Flux First Block Cache (FBC)** integration, and native **Hi-Res Fix** support.
11-
12-
Unlike standard browser wrappers, Resolver is built with a **Hybrid Architecture** (Capacitor 6.0 + Vanilla JS) and utilizes **Native Android Foreground Services** to ensure your generation queues never die in the background.
7+
**Resolver** is a high-performance, native Android interface for AI image generation. It is built using a **Hybrid Architecture** (Capacitor 6.0 + Vanilla JS) and utilizes **Native Android Foreground Services** to ensure generation queues remain active in the background or while the screen is off.
138

149
---
1510

@@ -66,108 +61,84 @@ Unlike standard browser wrappers, Resolver is built with a **Hybrid Architecture
6661

6762
## ⚡ Key Features
6863

69-
### ⚡ v1.3 Connectivity & Power
70-
* **Smart Connection Mode:** Toggle instantly between **Local Mode** (Auto-appends ports to local IPs) and **External Mode** (Enforces HTTPS for Ngrok/Cloud tunnels, ignoring ports).
71-
* **Custom Service Ports:** Define independent ports for your WebUI (Default: `7860`), LLM Server (`1234`), ComfyUI (`8188`), and Power Service (`5000`) inside the CFG tab.
72-
* **Remote Wake & Kill:** Use the "Bojro Dev Power" controls to send Wake-on-LAN signals or a **KILL!** signal to immediately halt the backend server from your phone.
64+
### Forge WebUI Core
65+
* **Primary Integration:** Resolver is natively optimized for [WebUI Forge Neo (Haoming02)](https://github.com/Haoming02/sd-webui-forge-classic/tree/neo) and the original [lllyasviel Forge WebUI](https://github.com/lllyasviel/stable-diffusion-webui-forge).
66+
* **Native API Communication:** Utilizes direct API endpoints for rapid state synchronization and command execution.
67+
* **Smart Connection Mode:** Toggle between **Local Mode** (automatic port appending) and **External Mode** (enforced HTTPS for Ngrok/Cloudflare).
68+
* **Remote Power Control:** Integrated "Bojro Dev Power" for Wake-on-LAN and a remote **KILL!** signal to stop the backend server instantly.
7369

74-
### ⚡ ComfyUI Engine
75-
* **Workflow Orchestration:** Load standard API-compatible ComfyUI `.json` workflows directly into the app.
76-
* **Dynamic UI Generation:** The engine automatically parses the workflow graph to build native sliders, dropdowns, and text inputs for samplers, checkpoints, and prompts.
77-
* **Live WebSocket Feedback:** Features real-time step monitoring and binary image previews delivered via persistent socket connections.
78-
* **Integrated Gallery:** Automatically archives final ComfyUI outputs into the main Resolver history for cross-engine metadata analysis.
70+
### ComfyUI Engine
71+
* **Workflow Orchestration:** Execute API-compatible ComfyUI `.json` workflows directly within the app.
72+
* **Dynamic UI Generation:** Automatically parses workflow graphs to generate native sliders, dropdowns, and text inputs for samplers, checkpoints, and prompts.
73+
* **Live WebSocket Feedback:** Real-time step monitoring and binary image previews via persistent socket connections.
74+
* **Unified Gallery:** Automatically archives ComfyUI outputs into the main history for cross-engine metadata analysis.
7975

80-
### ⚡ Flux First Block Cache (FBC)
81-
* **Native Integration:** First mobile client to support the **Forge Block Cache** extension.
82-
* **Speed Boost:** Skip redundant diffusion steps (`bnb-nf4`, `fp8`) to significantly reduce generation time on Flux models.
83-
* **Tunable:** Adjust threshold and skip steps directly from the Flux UI.
76+
### Flux First Block Cache (FBC)(Only lllyasviel Forge WebUI)
77+
* **Performance Boost:** Native support for the Forge Block Cache extension to reduce generation times.
78+
* **Optimization:** Skips redundant diffusion steps for `bnb-nf4` and `fp8` Flux models.
79+
* **Direct Control:** Threshold and skip step adjustments accessible directly via the Flux UI.
8480

85-
### ⚡ Native Hi-Res Fix
81+
### Native Hi-Res Fix
8682
* **Upscaling Engine:** Full support for `Hires. fix` across SDXL, Flux, and Qwen modes.
87-
* **Controls:** Select upscalers (ESRGAN, etc.), set denoising strength, and upscale factors within the app.
83+
* **Customization:** Adjustable upscalers (ESRGAN, etc.), denoising strength, and upscale factors.
8884

89-
### ⚡ True Background Generation
90-
* **Native Foreground Service:** Uses a Java-based service and Wake Locks to keep WebSocket connections alive during long Flux GGUF workflows or massive batch queues, even when the screen is off.
91-
* **Queue Persistence:** Robust state management system for Ongoing, Next, and Completed jobs.
85+
### True Background Generation
86+
* **Foreground Service:** Java-based service with Wake Locks to maintain active WebSocket connections during long Flux GGUF tasks or batch queues.
87+
* **Persistence:** Robust management for Ongoing, Next, and Completed jobs.
9288

93-
### ⚡ Advanced Generation Engines
94-
* **Flux GGUF Optimized:** Includes dedicated selectors for VAE, CLIP, and T5-XXL models, with support for specific quantization bits.
95-
* **Qwen / Z-Image Turbo:** Specialized "Turbo Generate" mode for Qwen models with dense narrative support.
96-
* **SDXL Powerhouse:** Full control over sampling, scheduling, and aspect ratio locking.
97-
* **Mobile Inpainting:** A touch-optimized canvas editor with mask blurring, soft inpainting, and denoising controls.
89+
### Advanced Generation Engines
90+
* **Flux GGUF:** Support for VAE, CLIP, and T5-XXL selectors with specific quantization bit settings.
91+
* **Qwen / Z-Image Turbo:** Specialized mode for Qwen models with dense narrative support.
92+
* **SDXL Powerhouse:** Comprehensive control over sampling, scheduling, and aspect ratio locking.
93+
* **Mobile Inpainting:** Touch-optimized canvas editor with mask blurring and denoising controls.
9894

99-
### ⚡ Magic Prompt (Local LLM)
100-
* **Smart Expansion:** Integrates with local LLM servers (LM Studio, Ollama) to translate simple ideas into professional prompts.
101-
* **Dynamic Mode Switching:** Automatically applies specialized system prompts for SDXL, Flux, and Qwen narratives.
95+
### Magic Prompt (Local LLM)
96+
* **Smart Expansion:** Connects to local LLM servers (LM Studio, Ollama) to translate simple ideas into complex prompts.
97+
* **Mode Awareness:** Automatically switches system instructions based on the active engine (SDXL, Flux, or Qwen).
10298

10399
---
104100

105101
## ⚡ Backend Setup (Mandatory)
106102

107-
To enable communication between the Android app and your PC, you must add the following launch arguments to your backend services:
108-
109103
**WebUI Forge:**
110-
Edit your `webui-user.bat` and set the following:
104+
In `webui-user.bat`, set:
111105
`set COMMANDLINE_ARGS=--listen --api --cors-allow-origins *`
112106

113107
**ComfyUI:**
114-
Append these flags to your launch command (after the python call):
108+
Append these flags to your launch command:
115109
`--listen --enable-cors-header *`
116110

117111
---
118112

119113
## ⚡ Magic Prompt (LLM) Setup
120114

121-
To use the "Magic Prompt" feature, it is highly recommended to host a **Bojro PromptMaster** model on your PC.
122-
123-
1. **Download a Model:** Get the GGUF version of your choice from Hugging Face:
124-
* 👉 **[PromptMaster v2 (Uncensored)](https://huggingface.co/bojrodev/BojroPromptMaster_uncensored_v2-8B)** (Recommended)
125-
* 👉 **[PromptMaster v1 (Uncensored)](https://huggingface.co/bojrodev/BojroPromptMaster-v1-8B)**
126-
2. **Run Server:** Load the model into **LM Studio** or **Ollama** and ensure it is serving on your local network (e.g., `http://192.168.1.10:1234`).
127-
3. **Connect App:**
128-
* In Resolver, tap the **Bot Icon** to open the Magic Prompt modal.
129-
* Enter your LLM Server URL (or configure the Port in the **CFG Tab**) and tap **Connect**.
130-
* The app will automatically use specialized system instructions for each image generation mode.
115+
Host a **Bojro PromptMaster** model on your PC:
116+
1. **Download:** [PromptMaster v2 (Uncensored)](https://huggingface.co/bojrodev/BojroPromptMaster_uncensored_v2-8B) or [v1](https://huggingface.co/bojrodev/BojroPromptMaster-v1-8B).
117+
2. **Server:** Load into **LM Studio** or **Ollama** on your local network.
118+
3. **Connect:** Open the Bot modal in Resolver, enter the server URL, and save.
131119

132120
---
133121

134-
## ⚡ Remote PC Wake (The Power Button)
135-
136-
Resolver features a remote power signal to start (or stop) your WebUI directly from the app home screen.
137-
138-
**Requirement:** Requires the `BojroPowerv3portable.exe` helper running on your PC.
122+
## ⚡ Remote PC Wake
139123

140-
1. **Run Helper:** Execute the helper app on your PC.
141-
2. **Configure:** Go to the **CFG Tab** in Resolver and set your **Wake Port** (Default: `5000`).
142-
3. **Wake:** Tap the Power Button (ტ) to send a start signal.
143-
4. **Kill:** Tap the **KILL!** button to remotely stop the services.
144-
145-
---
146-
147-
## ⚡ Workflow & LoRA Management
148-
* **Neo Bridge:** Browse LoRAs by folder with smart thumbnail caching.
149-
* **Config Injection:** Set preferred weights and trigger words that auto-inject into your prompts.
150-
* **Metadata Analysis:** Built-in PNG Info reader with one-tap "Copy to Mode" functionality to restore parameters from history.
124+
Requires `BojroPowerv_x_portable.exe` on the host PC.
125+
1. **Run Helper:** Execute the utility on your PC.
126+
2. **Configure:** Set the **Wake Port** (Default: `5000`) in the **CFG Tab**.
127+
3. **Operation:** Use the Power Button (ტ) to start or the **KILL!** button to halt services.
151128

152129
---
153130

154131
## ⚡ Installation & Building
155132

156133
### Installation
157-
1. Install **WebUI Forge (Neo)** and/or **ComfyUI** on your PC.
158-
2. Download the latest `.apk` from the **Releases Page**.
159-
3. Ensure your phone and PC are on the same Wi-Fi.
160-
4. Navigate to the **CFG Tab**, enter your PC's IP, and tap **Save Configuration**.
134+
1. Install **WebUI Forge** and/or **ComfyUI** on your PC.
135+
2. Download the latest `.apk` from the **Releases**.
136+
3. Ensure phone and PC are on the same Wi-Fi.
137+
4. Set PC IP in the **CFG Tab** and save.
161138

162139
### Building from Source
163140
```bash
164-
# Clone the repository
165141
git clone [https://github.com/bojrodev/Resolver-WebUI-Forge-Client.git](https://github.com/bojrodev/Resolver-WebUI-Forge-Client.git)
166-
167-
# Install dependencies
168142
npm install
169-
170-
# Sync and open Android project
171143
npx cap sync
172144
npx cap open android
173-
# Open the Android Project in Android Studio

0 commit comments

Comments
 (0)