DEV Community

Daniel Igel
Daniel Igel

Posted on

5 text analysis tasks in one POST — no LLM key needed

#ai #api #llm #nlp

Setting up NLP in 2026 still means choosing between expensive ML platforms, fragile Python environments, or stitching together multiple services — each requiring its own LLM key.

One POST returns sentiment, a summary, named entities, keyword extraction, or zero-shot classification — all powered by Gemini, no LLM key of your own needed:

curl --request POST \
  --url 'https://ai-text-analysis-api.p.rapidapi.com/api/v1/sentiment' \
  --header 'x-rapidapi-key: YOUR_RAPIDAPI_KEY' \
  --header 'x-rapidapi-host: ai-text-analysis-api.p.rapidapi.com' \
  --header 'content-type: application/json' \
  --data '{"text": "The new product update is absolutely fantastic!"}'
Enter fullscreen mode Exit fullscreen mode

Response:

{ "sentiment": "positive", "score": 0.95, "explanation": "Strong positive language." }
Enter fullscreen mode Exit fullscreen mode

Five dedicated endpoints: /sentiment, /summarize, /entities, /classify, /keywords. Need a 3-sentence summary? Send {"text": "...", "sentences": 3} and get back summary, sentences[], and wordCount. Zero-shot classification takes any categories array — no model training or fine-tuning required.

const res = await fetch('https://ai-text-analysis-api.p.rapidapi.com/api/v1/entities', {
  method: 'POST',
  headers: {
    'x-rapidapi-key': process.env.RAPIDAPI_KEY,
    'x-rapidapi-host': 'ai-text-analysis-api.p.rapidapi.com',
    'content-type': 'application/json',
  },
  body: JSON.stringify({ text: 'Tim Cook announced the iPhone in Cupertino in 2007.' }),
});
const { entities, count } = await res.json();
// entities: [{ text: 'Tim Cook', type: 'PERSON' }, { text: 'Cupertino', type: 'LOCATION' }, ...]
Enter fullscreen mode Exit fullscreen mode

Accepts up to 16,000 characters per request. Works from any HTTP client — Node, Deno, Cloudflare Workers, plain browser fetch. No SDK, no Python environment to set up.

Free tier on RapidAPI: https://rapidapi.com/danieligel/api/ai-text-analysis-api

Which text analysis task do you reach for most often in a backend project — sentiment, entity extraction, or something else?

Top comments (0)