How does Lime Chat work?

Lime Chat uses a local AI model that runs entirely in your browser via WebGPU. When you upload a PDF, it's indexed locally. When you ask questions, the AI finds relevant passages and generates answers based on the document content.

Is my data sent to any servers?

No, everything runs locally. The AI model is downloaded once and runs in your browser. Your documents and conversations never leave your device.

What are the system requirements?

You need a modern browser with WebGPU support (Chrome 113+, Edge 113+). A dedicated GPU with 4-8GB VRAM is recommended for best performance, though integrated graphics can work with smaller models.

How long does the model take to load?

The first load downloads the model (1-4GB depending on model size), which may take a few minutes. Subsequent loads are much faster as the model is cached in your browser.

Can I chat with multiple documents at once?

Yes! You can upload multiple PDFs and select which ones to include in your chat. The AI will search across all selected documents when answering questions.

What types of questions can I ask?

You can ask any questions about your documents - summaries, specific facts, comparisons, explanations, and more. The AI will cite the relevant page numbers in its answers.

Back to tools

Lime ChatLocal AI

Chat with your PDFs using AI that runs entirely on your device. No data leaves your browser.

100% Local AI Processing

AI Model

Checking WebGPU support...

Documents

Drop PDF or click

No documents indexed yet

Start a conversation

Load an AI model to start chatting

How Lime Chat works

1.Load an AI model - runs entirely in your browser using WebGPU
2.Upload PDFs - they're indexed locally for semantic search
3.Ask questions - the AI finds relevant passages and generates answers
4.Everything stays private - no data ever leaves your device