Files changed (1) hide show
  1. README.md +0 -143
README.md DELETED
@@ -1,143 +0,0 @@
1
- ---
2
- title: AnyCoder
3
- emoji: 🚀
4
- colorFrom: blue
5
- colorTo: purple
6
- sdk: docker
7
- app_port: 7860
8
- pinned: false
9
- disable_embedding: false
10
- hf_oauth: true
11
- hf_oauth_scopes:
12
- - manage-repos
13
- ---
14
-
15
- > **Note:** This is the Docker Space configuration for the React frontend version.
16
- > For the original Gradio app, see `README_GRADIO.md`.
17
-
18
- # AnyCoder - AI Code Generator with React Frontend
19
-
20
- AnyCoder is a full-stack AI-powered code generator with a modern React/TypeScript frontend and FastAPI backend. Generate applications by describing them in plain English, with support for multiple AI models and one-click deployment to Hugging Face Spaces.
21
-
22
- ## 🎨 Features
23
-
24
- - **Modern React UI**: Apple-inspired design with VS Code layout
25
- - **Real-time Streaming**: Server-Sent Events for live code generation
26
- - **Multi-Model Support**: MiniMax M2, DeepSeek V3, and more via HuggingFace InferenceClient
27
- - **Multiple Languages**: HTML, Gradio, Streamlit, React, Transformers.js, ComfyUI
28
- - **Authentication**: HuggingFace OAuth + Dev mode for local testing
29
- - **One-Click Deployment**: Deploy generated apps directly to HF Spaces
30
-
31
- ## 🏗️ Architecture
32
-
33
- ```
34
- anycoder/
35
- ├── backend_api.py # FastAPI backend with streaming
36
- ├── frontend/ # Next.js React frontend
37
- │ ├── src/
38
- │ │ ├── app/ # Pages (page.tsx, layout.tsx, globals.css)
39
- │ │ ├── components/ # React components
40
- │ │ ├── lib/ # API client, auth utilities
41
- │ │ └── types/ # TypeScript types
42
- │ └── package.json
43
- ├── anycoder_app/ # Original Gradio app modules
44
- │ ├── agent.py
45
- │ ├── config.py
46
- │ ├── deploy.py
47
- │ └── ...
48
- ├── app.py # Original Gradio interface
49
- ├── requirements.txt # Python dependencies
50
- ├── Dockerfile # Docker Space configuration
51
- └── start_fullstack.sh # Local development script
52
- ```
53
-
54
- ## 🚀 Quick Start
55
-
56
- ### Local Development
57
-
58
- 1. **Backend**:
59
- ```bash
60
- export HF_TOKEN="your_huggingface_token"
61
- export GEMINI_API_KEY="your_gemini_api_key"
62
- python backend_api.py
63
- ```
64
-
65
- 2. **Frontend** (new terminal):
66
- ```bash
67
- cd frontend
68
- npm install
69
- npm run dev
70
- ```
71
-
72
- 3. Open `http://localhost:3000`
73
-
74
- ### Using start script:
75
- ```bash
76
- export HF_TOKEN="your_token"
77
- export GEMINI_API_KEY="your_gemini_api_key"
78
- ./start_fullstack.sh
79
- ```
80
-
81
- ## 🐳 Docker Space Deployment
82
-
83
- This app runs as a Docker Space on HuggingFace. The Dockerfile:
84
- - Builds the Next.js frontend
85
- - Runs FastAPI backend on port 7860
86
- - Uses proper user permissions (UID 1000)
87
- - Handles environment variables securely
88
-
89
- ## 🔑 Authentication
90
-
91
- - **Dev Mode** (localhost): Mock login for testing
92
- - **Production**: HuggingFace OAuth with manage-repos scope
93
-
94
- ## 📝 Supported Languages
95
-
96
- - `html` - Static HTML pages
97
- - `gradio` - Python Gradio apps
98
- - `streamlit` - Python Streamlit apps
99
- - `react` - React/Next.js apps
100
- - `transformers.js` - Browser ML apps
101
- - `comfyui` - ComfyUI workflows
102
-
103
- ## 🤖 Available Models
104
-
105
- - **Gemini 3 Pro Preview** (Default) - Google's latest with deep thinking & Google Search
106
- - MiniMax M2 (via HF router with Novita)
107
- - DeepSeek V3/V3.1
108
- - DeepSeek R1
109
- - And more via HuggingFace InferenceClient
110
-
111
- ## 🎯 Usage
112
-
113
- 1. Sign in with HuggingFace (or use Dev Login locally)
114
- 2. Select a language and AI model
115
- 3. Describe your app in the chat
116
- 4. Watch code generate in real-time
117
- 5. Click **🚀 Deploy** to publish to HF Spaces
118
-
119
- ## 🛠️ Environment Variables
120
-
121
- - `HF_TOKEN` - HuggingFace API token (required)
122
- - `GEMINI_API_KEY` - Google Gemini API key (required for Gemini 3 Pro Preview)
123
- - `POE_API_KEY` - Poe API key (optional, for GPT-5 and Claude models)
124
- - `DASHSCOPE_API_KEY` - DashScope API key (optional, for Qwen models)
125
- - `OPENROUTER_API_KEY` - OpenRouter API key (optional, for Sherlock models)
126
- - `MISTRAL_API_KEY` - Mistral API key (optional, for Mistral models)
127
-
128
- ## 📦 Tech Stack
129
-
130
- **Frontend:**
131
- - Next.js 14
132
- - TypeScript
133
- - Tailwind CSS
134
- - Monaco Editor
135
-
136
- **Backend:**
137
- - FastAPI
138
- - HuggingFace Hub
139
- - Server-Sent Events (SSE)
140
-
141
- ## 📄 License
142
-
143
- MIT