Text understanding  /  API ~

TextMind

Seven zero-cost deterministic endpoints (keywords, entities, sentiment, language, readability) plus four neural ones running on Workers AI, all behind one key and one schema.

checking 9 endpoints Cloudflare edge
ai-text · live
$ curl -s -X POST "https://textmind.ayamk.in/v1/sentiment" -H "content-type: application/json" -d '{"text":"The deploy went smoothly and the team is thrilled."}' | jq
{
  "text": "The deploy went smoothly and the team is thrilled.",
  "score": 0.62,
  "comparative": 0.0775,
  "label": "positive",
  "tokens": 8,
  "sentences": [
    { "text": "The deploy went smoothly and the team is thrilled.", "score": 0.62 }
  ],
  ...
01 - ENDPOINTS

9 endpoints, one key

Every route below is described in the OpenAPI document. Import it into Postman or generate a client.

MethodPathWhat it does
GET/v1/keywordsTF-IDF style ranking with 1-3 grams and density.
POST/v1/entitiesEmails, URLs, phones, dates, money, hashtags, handles, IDs.
POST/v1/sentimentLexicon sentiment with per-sentence breakdown.
POST/v1/detect-language12 languages, confidence and runners up.
POST/v1/readabilityFlesch, grade level, sentence statistics.
POST/v1/summarizeNeural summary, paragraph or bullets.
POST/v1/rewriteRewrite in a chosen tone.
POST/v1/classifyZero-shot classification against your labels.
POST/v1/translateNeural translation between 100 languages.
02 - RESPONSE

What you actually get back

curl
curl -s -X POST "https://textmind.ayamk.in/v1/sentiment" -H "content-type: application/json" -d '{"text":"The deploy went smoothly and the team is thrilled."}' | jq
200 application/json
{
  "text": "The deploy went smoothly and the team is thrilled.",
  "score": 0.62,
  "comparative": 0.0775,
  "label": "positive",
  "tokens": 8,
  "sentences": [
    { "text": "The deploy went smoothly and the team is thrilled.", "score": 0.62 }
  ],
  "topPositive": ["smoothly", "thrilled"],
  "topNegative": [],
  "meta": { "costClass": "free", "durationMs": 2 }
}
Limit worth knowing. Neural endpoints depend on Cloudflare Workers AI availability. Output quality is model-dependent; treat summaries as drafts, not facts.
03 - WHY

Built for the boring part

Free endpoints are actually free

The deterministic side is pure code on the edge: no model, no tokens, no per-call cost. Run your keywords and entities extraction over a whole archive.

Neural where it earns its keep

Summaries, rewrites, zero-shot labels and translation go through Workers AI with tight max_tokens caps, so latency stays in the seconds and cost stays predictable.

Same response shape

Every endpoint returns a meta block with costClass free or neural, model id and duration, so you can route traffic and forecast spend.

04 - PRICING

Priced per request, not per seat

Basic
$0
25 requests / month
Card required by the marketplace. Enough to ship an integration.
Ultra
$75/mo
100,000 requests
Batch jobs and backfills.
Mega
$150/mo
400,000 requests
For platform-scale volume. Overage billed per request.

All plans are managed by the RapidAPI marketplace, which applies a marketplace fee before payout. Annual and custom limits are available on request.

05 - USE CASES

Where this earns its place

01

Support ticket triage

02

Content moderation queues

03

Bulk article tagging

04

Localisation drafts

06 - FAQ

Questions we get asked

Which endpoints cost nothing?

keywords, entities, sentiment, detect-language and readability are deterministic code. They are marked costClass free in every response.

What happens if the AI model is unavailable?

You get a 503 with code AI_UNAVAILABLE and a retry hint. The free endpoints keep working, so your product degrades instead of breaking.

Can I choose the model?

A default is pinned per endpoint. Ask for a specific model and it can be wired into your plan.

Is my text stored?

No. Text is processed in memory inside the request and discarded.

Subscribe to TextMind

Free tier included. The listing is published on the RapidAPI marketplace; the key works against every endpoint on this page.