Sentiment Analyzer
Analyze text sentiment with AI
About Sentiment Analyzer
Sentiment Analyzer reads a piece of text, such as a customer review, a comment, or a support message, and says whether its overall tone is Positive, Negative, or Neutral. The answer, written by OpenAI's GPT-4o mini, also gives a confidence percentage and the phrases that point to that sentiment. The percentage is the model's own estimate, not a measured accuracy. Your text goes through our server to OpenAI. One text per request, up to 8,000 characters; there is no batch mode and no spreadsheet import.
How to analyze the sentiment of a text
- 1Paste one review or message into the "Paste text to analyze sentiment…" box.
- 2Click "Analyze sentiment".
- 3Read the verdict, the confidence percentage, and the explanation under "Result".
- 4Click "Copy" or "Download" (sentiment.txt) to keep the analysis.
Formats and limits
- Input
- One review or message, up to 8,000 characters
- Output
- Positive, Negative, or Neutral, a confidence percentage, and the key indicators, in plain text
- Batch
- No; one text per request
- Usage limits
- Hourly and daily request limits per connection, shared with the site's other tools that rely on a paid service, and a monthly site budget
Where your text is processed
Your text goes to our server and from there to OpenAI's GPT-4o mini, which returns the analysis. Each review is a separate request, and nothing from one analysis is kept for the next: we store neither the text nor the result. When you analyze customer messages, remove names and contact details first.
Frequently asked questions
Is the confidence percentage a measured accuracy?
No. It is the model's own estimate of how sure it is, written into its answer. The tool does not compute it, so two runs on the same text can give different figures.
Can I analyze many reviews at once?
No. The page sends one text per request. If you paste several reviews together, you get one overall verdict for all of them; paste them one at a time to classify each review.
Does it understand sarcasm?
No, not reliably. Irony, sarcasm, and mixed reviews, such as praise for a product with complaints about delivery, can be misread. Read the explanation in the result to see which words led to the verdict.
Which languages can I analyze?
Any language you paste: the page neither sets nor blocks one, and the text goes to the model as it is. The site has not measured accuracy by language, so read results in languages other than English with extra care.
Tips and tricks
- ✓Split a review that praises one thing and criticizes another into two parts to see each sentiment separately.