# FluentAI

> FluentAI is a free, privacy-first AI chat client that runs 100+ LLM models offline on your device. Available on Android with desktop versions coming soon. Supports local models (Llama 3, Gemma 3, DeepSeek, Mistral, Phi, Qwen) and cloud providers (Claude, GPT-4, Gemini, OpenRouter).

FluentAI is built for users who want powerful AI without sacrificing privacy. All local conversations stay on-device with zero data collection, zero telemetry, and zero tracking. Cloud providers are optional and use your own API keys.

---

## Product Overview

FluentAI is a cross-platform AI chat application that puts privacy first. It allows users to:

- Run AI models entirely on their device with no internet connection
- Chat with 100+ different AI models from various providers
- Use voice chat with 5 different conversation modes
- Upload documents and create knowledge bases with on-device RAG
- Use built-in tools and connect to external services via MCP (Model Context Protocol)
- Organize conversations with folders, tags, pinning, and branching
- Export chats in multiple formats including audio podcasts
- Create custom AI personas with unique system prompts

The app is completely free for local model usage. Cloud providers require the user's own API keys.

---

## Features

### Privacy First
Your conversations never leave your device. No data collection, no tracking, no cloud required. All local model inference happens on-device. FluentAI has zero telemetry, zero tracking, and zero data collection. Users can audit the open source code themselves for full transparency.

### 100+ AI Models
Run Llama 3, Gemma 3, DeepSeek, Mistral, Phi, and Qwen locally on your device using GGUF format models. Or connect to cloud providers including Claude (Anthropic), GPT-4 (OpenAI), Gemini (Google), and 200+ models via OpenRouter. Ollama local server is also supported as infrastructure.

### Voice Chat
Talk to AI naturally with 5 conversation modes:
- Normal: Standard voice conversation
- Interview: AI interviews you on a topic
- Learning: AI teaches you interactively
- Storytelling: AI tells stories with voice
- Translation: Real-time voice translation

### Knowledge Bases
Upload PDFs, text files, and documents to create knowledge bases. FluentAI uses RAG (Retrieval-Augmented Generation) to search and answer questions from your files — all processed on-device with semantic search.

### Built-in Tools & MCP
FluentAI comes with built-in tools: calculator, web search, weather, date/time, and AI memory. Plus full Model Context Protocol (MCP) support to connect to GitHub, Slack, Notion, and 20+ other services.

### Rich Content & Code
Beautiful syntax-highlighted code blocks, LaTeX math rendering, HTML/SVG previews, and full Markdown support. Perfect for developers, students, and researchers.

### Templates & AI Personas
Choose from built-in prompt templates or create your own. Set up custom AI personas with unique system prompts — from a coding assistant to a creative writing partner. Supports auto-fill for quick template use.

### Bring Your Own Model
Import any GGUF model file or load models directly from Hugging Face. FluentAI doesn't lock you into a curated list — use any compatible model you want for total freedom and flexibility.

### Chat Organization
Folders, tags, pinning, branching, and search. Keep your conversations organized your way.

### Export & Share
Export chats as text, Markdown, JSON, or even as audio podcasts. Share conversations anywhere.

### Completely Free
No $20/month subscriptions. Use powerful local models at zero cost, forever. Cloud providers only require your own API keys.

### Cross-Platform
Available on Android today via Google Play, F-Droid, and IzzyOnDroid. iOS, Windows, macOS, Linux, and Web versions are coming soon.

---

## Supported Models

### Local / On-Device Models (GGUF format, runs offline)
| Model | Provider | Category |
|-------|----------|----------|
| Llama 3 | Meta | On-device |
| Gemma 3 | Google | On-device |
| DeepSeek | DeepSeek AI | On-device |
| Mistral | Mistral AI | On-device |
| Phi | Microsoft | On-device |
| Qwen | Alibaba Cloud | On-device |

### Cloud Providers (requires API key)
| Model | Provider | Category |
|-------|----------|----------|
| Claude | Anthropic | Cloud |
| GPT-4 | OpenAI | Cloud |
| Gemini | Google | Cloud |
| OpenRouter | Various (200+ models) | Cloud |

### Infrastructure
| Service | Description |
|---------|-------------|
| Ollama | Local server for running models |

---

## Comparison: FluentAI vs Competitors

| Feature | FluentAI | ChatGPT | Claude | Gemini |
|---------|----------|---------|--------|--------|
| Price | Free (local models) | Free / $20/mo | Free / $20/mo | Free / $20/mo |
| Privacy | On-device, zero collection | Cloud, data used for training | Cloud-based | Cloud, data used for training |
| Offline Mode | Yes | No | No | No |
| Model Choice | 100+ models | GPT-4 only | Claude only | Gemini only |
| Voice Chat | Yes | Paid | No | Yes |
| Open Source | Yes | No | No | No |

---

## Frequently Asked Questions

### Is FluentAI really free?
Yes! Local models are unlimited and completely free. If you want to use cloud providers like Claude, GPT-4, or Gemini, you just need your own API keys.

### Does it work offline?
Yes. Local GGUF models run entirely on your device with no internet connection needed. Download a model once and chat anywhere — on a plane, in remote areas, or simply off the grid.

### Which AI models are supported?
100+ models including Llama 3, Gemma 3, DeepSeek, Mistral, Phi, and Qwen locally. Plus Claude (Anthropic), GPT-4 (OpenAI), Gemini (Google), and 200+ models via OpenRouter in the cloud.

### Is my data private?
Absolutely. All local conversations stay on your device. No data is ever sent to our servers. We have zero telemetry, zero tracking, and zero data collection.

### What platforms are supported?
Android is available now on Google Play, F-Droid, and IzzyOnDroid. iOS, Windows, macOS, Linux, and Web versions are coming soon.

### How do I get started?
Download FluentAI from Google Play, pick a model (or download one for offline use), and start chatting. No account required.

---

## Pricing

FluentAI offers two tiers:

### Free (forever)
- 100+ local AI models (Llama, Gemma, DeepSeek, Mistral, Phi, Qwen)
- Unlimited offline chat — no internet needed
- Cloud providers with your own API keys (Claude, GPT-4, Gemini)
- Voice chat with 5 conversation modes
- Knowledge bases & on-device RAG
- Tool calling & MCP support
- Chat folders, tags, pinning, branching & search
- Export as text, Markdown & JSON
- 3 built-in themes
- Full Markdown, code highlighting & LaTeX

### Premium ($3.49 one-time purchase)
- Everything in Free, plus:
- Ad-free experience
- 9 premium themes (Sunset Orange, Cyber Neon, Arctic Ice & more)
- Cloud usage dashboard — track tokens, costs & usage patterns
- Advanced model settings (Top-P, frequency/presence penalties, per-chat overrides)
- Rich PDF & image export with syntax highlighting
- Advanced analytics — detailed usage stats & insights
- Unlimited performance benchmarks
- Priority access to new features
- Premium badge

### Try Before You Buy
Every premium feature can be trialed for free by watching a short ad. Get 24-hour access to premium themes and cloud dashboard, 1-hour access to advanced model settings, or unlock PDF export for individual chats — all without paying.

---

## Contact

- Email: readheights@gmail.com
- Google Play: https://play.google.com/store/apps/details?id=com.readheights.fluentai
- Developed by: Readheights Technologies Private Limited (https://readheights.com)

---

## Download Links

- Android (Google Play): https://play.google.com/store/apps/details?id=com.readheights.fluentai
- Android (F-Droid): Coming soon
- Android (IzzyOnDroid): Coming soon
- Mac (Apple Silicon): Coming soon
- Mac (Intel): Coming soon
- Windows: Coming soon
- Linux: Coming soon

---

## Technical Details

- Local models use GGUF format for on-device inference
- RAG (Retrieval-Augmented Generation) for document Q&A runs entirely on-device
- MCP (Model Context Protocol) support for connecting to external tools and services
- Built with Flutter for cross-platform support
- Static website built with Next.js, hosted on Vercel