Sentiment
Hosted LLM sentiment classification — handles sarcasm, mixed tone, and domain language the in-browser heuristic misses. Returns positive / negative / neutral. Uses 1 credit per run.
About Sentiment
Sentiment classifies text as positive, negative, or neutral using a hosted LLM that reads tone the way a person would, catching sarcasm, mixed feelings, and domain-specific language that simple keyword heuristics miss. It's the tool to reach for when you need a reliable read on how something is worded rather than a brittle count of "good" and "bad" words. It returns structured JSON and costs 1 credit per run.
- Category
- text
- Input
- Accepts: text/plain.
- Output
- Outputs: application/json.
- Cost
- Credit-metered
- Memory
- low
Common uses
- Score a batch of customer reviews where sarcasm flips the apparent meaning
- Gauge the tone of a support ticket before routing it to the right queue
- Classify survey free-text responses that mix praise and complaints in one breath
- Check whether a social mention is genuinely negative or just dryly worded
- Triage feedback in a domain where jargon confuses simple keyword scoring
- Add a sentiment label to chained text output for downstream filtering
Frequently asked questions
What does it return?
JSON (application/json) with a sentiment classification of positive, negative, or neutral for the input text.
How is this better than the free heuristic version?
The hosted LLM understands sarcasm, mixed tone, and domain language, whereas keyword-based heuristics often mislabel anything more subtle than plain praise or complaint.
Why does it cost a credit?
It runs on a hosted model rather than in your browser, so each classification uses 1 credit.
Is my text uploaded?
Yes. Because the classification comes from a hosted model, your text is sent to it for processing, unlike the free in-browser tools.
What input format does it take?
Plain text (text/plain). Paste or chain in the content you want classified.
Keywords
- sentiment
- tone
- classification
- pro
- llm
- hosted