x10 is an interactive, real-time AI multi-model interface that lets you broadcast prompts to 10 AI models in parallel, chat with individual models, and watch them engage in live debates inside the Debate Chamber.
It is designed with a premium, responsive user interface optimized for both desktop grid views and mobile horizontal swipable chats.
- Desktop Grid: Send a single prompt and watch all 10 models generate streaming responses side-by-side in real-time.
- Mobile Swipe View: Swipe horizontally left-to-right to compare model responses in parallel on mobile viewports.
- Single Model Focus: Isolate any of the 10 models for a dedicated chat experience.
- Vision / Image Uploads: Select models like Gemini 2.0 or Llama 4 Scout to upload images (PNG/JPEG) and ask visual prompts.
- Live AI Debate: Convene a turn-based live debate among the models on any topic.
- Scribe's Ledger: A dedicated "Scribe" compiles and updates a live summary of key arguments in real-time.
- User Intervention: Inject your own statements into the live debate at any turn.
- Final Verdict: The debate automatically concludes after 20 turns (or manually) and streams a structured consensus verdict.
| Model Name | API Provider / Backend | Vision Support |
|---|---|---|
| Llama 3.3 | Groq API Key 1 | ❌ |
| Qwen 3.6 | Groq API Key 1 | ❌ |
| ALLaM 7B | Groq API Key 1 | ❌ |
| Qwen 3 | Groq API Key 2 | ❌ |
| Compound | Groq API Key 2 | ❌ |
| GPT-OSS | Groq API Key 2 | ❌ |
| Llama 4 Scout | Groq API Key 2 | ✅ |
| Gemini 2.0 | Google Gemini API | ✅ |
| GPT-4o | Pollinations API (Stateless) | ❌ |
| Llama 3.2 | Cloudflare Workers AI | ❌ |
Ensure you have Python 3.8+ installed. Navigate to the project root and install requirements:
pip install -r requirements.txtCreate a file named .env in the root folder and add your API credentials:
GROQ_API_KEY_1=your_groq_key_1
GROQ_API_KEY_2=your_groq_key_2
GOOGLE_API_KEY=your_gemini_key
CLOUDFLARE_ACCOUNT_ID=your_cloudflare_account_id
CLOUDFLARE_API_TOKEN=your_cloudflare_api_tokenLaunch the FastAPI development server:
uvicorn main:app --reloadOpen http://127.0.0.1:8000 in your browser.
The repository is configured for deployment on the Vercel platform using @vercel/python:
- Import the repository into your Vercel Dashboard.
- Configure your Environment Variables in the project settings.
- Click Deploy. Vercel will automatically read
vercel.jsonand configure serverless routing.
- Created by: Afroz Alam
- Contact Email: [email protected]