Skip to content
Ad: top-banner (320x50)

Sentiment Analyzer

Instantly detect whether text is positive or negative using a small AI model that runs entirely in your browser. Analyze product reviews, feedback forms, or your own drafts before sending.

Your data never leaves your device
Ad: sidebar (300x250)
Ad: mid-banner (300x250)

Frequently Asked Questions

  • What does "sentiment" mean?

    Sentiment is a label that describes whether the emotional tone of a piece of text leans positive (happy, satisfied, enthusiastic) or negative (angry, disappointed, critical). This tool assigns a label and a confidence score to each line or paragraph.

  • Which model is used?

    A DistilBERT model fine-tuned on the Stanford Sentiment Treebank (SST-2), served as an ONNX file (~67 MB, quantized). It runs via @huggingface/transformers entirely in your browser and is cached after first use.

  • Is my text uploaded anywhere?

    No. All inference runs locally in your browser via WebAssembly. Your text never leaves your device.

  • How accurate is it?

    The model performs well on English product reviews, tweets, and short messages. It can struggle with sarcasm, mixed sentiment, and domain-specific jargon (e.g. medical or legal text). Always sanity-check the result.