R97 AI is a premium, client-side intelligence platform designed by Rehan Ahmad. It executes purely in the browser environment, connecting directly to NVIDIA's Llama 3.1 inference engine for complex reasoning and deep context resolution.
- Deep Reasoning: Powered by NVIDIA NIM (Meta Llama 3.1) for complex problem solving and detailed text responses directly through client-side requests.
- Client-Side Storage: Fully local history and chat states stored safely in
localStorage.
- Native Tongues: Advanced support for English, Hindi, and Hinglish.
- Style Sync: The AI automatically adopts the language, tone, and slang of the user for a truly native feel.
- In-Browser TTS: Pure client-side Web Speech API Text-to-Speech synthesis for natural voice readout without backend computation.
- Smart Formatting: Automatically filters noise (URLs, code blocks) during voice synthesis for a cleaner experience.
- Security: Built-in 100% Client-Side architecture guarantees no permanent storage outside the native environment.
- Frontend: React 18, Vite, Tailwind CSS, Framer Motion.
- Backend / Deployment: Removed. Now purely a Serverless Client-Side Application (SPA).
- Models: NVIDIA Llama 3.1 NIM directly queried via client fetch.
- Styling: Cyber-Pixel aesthetic with custom-tuned HUD elements.
R97 AI is the signature project of Rehan Ahmad.
- GitHub: Ft976
- LinkedIn: Rehan Ahmad
For detailed information on the usage, privacy, and development of R97 AI, please refer to:
- License - MIT License information.
- Privacy Policy - How we handle your data.
- Terms of Service - Guidelines for usage and AI behavior.
- Contributing - How to help improve the project.
- Security - Reporting vulnerabilities.
Since the entire backend is removed, this runs entirely in the browser. You MUST configure the following Client-Side Environment variable:
NVIDIA_API_KEY=your_key_hereCopyright © 2026 Rehan Ahmad. All Rights Reserved.