Vantage is a cutting-edge AI editorial assistant designed to transform Google Docs into dynamic, intelligent workspaces. Leveraging the power of Google's Gemini 3.0 and 2.5 models, Vantage acts as a proactive collaborator, offering deep structural rewrites, real-time voice discussions, fact-checking, and instant visual asset generation.
- 🧠 Deep Polish (Gemini 3.0 Pro): Utilizes advanced "Thinking" capabilities to analyze document structure and tone, offering world-class rewrites and stylistic critiques.
- 🎨 Nano Banana (Visual Intelligence): Context-aware image suggestions powered by Gemini 3.0 Pro, instantly generated using Gemini 2.5 Flash Image.
- 🎙️ Voice Review (Live API): Talk to your document in real-time. A low-latency, multimodal voice interface allows for hands-free brainstorming and summarization.
- ✅ Fact Checking (Grounding): Verifies claims against Google Search to ensure accuracy and provides source links.
- 💬 Smart Collaboration: Suggests resolutions for existing comments and proactively recommends new improvements for clarity and impact.
Vantage is built with a modern frontend stack and direct integration with Google's GenAI SDK.
- Frontend Framework: React 19 with TypeScript.
- Styling: Tailwind CSS for a responsive, beautiful UI.
- AI SDK:
@google/genaifor direct interaction with Gemini models. - Audio Processing: Web Audio API for real-time PCM streaming to the Gemini Live API.
Vantage matches the specific capabilities of Gemini models to the user's task:
| Feature | Model | Reason |
|---|---|---|
| Chat & Basic Logic | gemini-2.5-flash |
Low latency, high throughput for standard interactions. |
| Deep Polish | gemini-3-pro-preview |
Utilizes the Thinking budget (16k tokens) for complex reasoning and editorial judgment. |
| Visual Context | gemini-3-pro-preview |
Understands document nuance to write highly detailed image prompts. |
| Image Generation | gemini-2.5-flash-image |
Fast, high-quality image generation (Nano Banana) for visual assets. |
| Voice Mode | gemini-2.5-flash-native-audio |
Native audio-to-audio streaming for ultra-low latency conversations. |
- Node.js (v18+)
- A Google Cloud Project with the Gemini API enabled.
- An API Key from Google AI Studio.
-
Clone the repository
git clone https://github.com/yourusername/vantage.git cd vantage -
Install dependencies
npm install
-
Configure Environment Create a
.envfile in the root directory and add your API key. (Note: In a production environment, handle keys securely via backend proxies).API_KEY=your_gemini_api_key_here
-
Run the Development Server
npm run dev
-
Open in Browser Navigate to
http://localhost:5173.
Note: For the "Talk to Doc" feature, ensure you grant microphone permissions when prompted.
We welcome contributions to make Vantage even smarter!
- Fork the repository.
- Create a Feature Branch (
git checkout -b feature/AmazingFeature). - Commit your changes (
git commit -m 'Add some AmazingFeature'). - Push to the branch (
git push origin feature/AmazingFeature). - Open a Pull Request.
Distributed under the MIT License.