Route form submissions, detect spam, triage support tickets, qualify leads - with one API call. Plugs into n8n, Zapier, Make, or any HTTP workflow in minutes.
# Classify an inbound contact form submission
curl https://api.classifaily.com/v1/classify \
-H "Authorization: Bearer cai_live_..." \
-H "Content-Type: application/json" \
-d '{
"input": "Get 10,000 followers fast! Click here now!!!",
"type": "text",
"categories": ["genuine_inquiry","spam","abuse","other"]
}'
// Response - block or route in your workflow
{
"status": "completed",
"result": {
"label": "spam",
"confidence": 0.97
},
"meta": { "latency_ms": 183 }
}Works with your stack and your automation tools
Generate a key from your dashboard and add it to your request header - or drop it into an n8n HTTP node, a Zapier webhook, or a Make module in seconds.
Pass categories inline, reference a saved schema, or let classifaily label automatically. Categories can be anything: spam / genuine, sales / support / billing, urgent / normal / low.
Get a clean JSON label and confidence score. Route it to a CRM, block a form submission, trigger a Slack alert, or branch your workflow - the logic is yours.
Classify form submissions, tickets, emails, documents, and feedback into any category set you define.
Send images by URL or base64. Label for content moderation, product categorization, or custom taxonomy.
Classify audio clips by content, tone, or topic. Ideal for call center routing and podcast tagging pipelines.
Submit video URLs for scene or content classification - great for content libraries and moderation pipelines.
Save category sets as named schemas. Reference a schema_id across requests to keep classification consistent at scale.
Native REST JSON means zero friction in n8n, Zapier, Make, or Pipedream. Drop in an HTTP node, wire up the result, done.
Register a webhook URL to receive classification results the moment they complete - ideal for async batch processing.
Median latency under 300ms - fast enough to classify form submissions synchronously before returning a response to the user.
From a single form field to millions of records - classifaily handles the decision so your workflow doesn't have to.
Classify contact form, signup, and comment submissions as genuine, spam, or abusive before they hit your inbox or database. Block or quarantine in real time.
Automatically determine intent behind a form submission - sales inquiry, support request, partnership, press - and route it to the right person or queue without manual reading.
Add a classification step to any n8n, Zapier, Make, or Pipedream workflow. One HTTP call returns a label your workflow can branch on, filter, or write to a field.
Tag and route incoming tickets to the right team - billing, technical, account, general - the moment they arrive. Cut triage time to zero.
Classify inbound leads by intent, company size signals, or deal urgency. Trigger different CRM sequences, Slack alerts, or rep assignments without manual review.
Classify inbound emails by type, urgency, or topic. Auto-assign, auto-label, or auto-respond based on the result - works alongside any email automation platform.
Flag user-generated content as safe, NSFW, spam, or abusive before it reaches your platform. Works across text, images, and comments at scale.
Classify NPS responses, reviews, and open-ended survey answers into themes automatically. No more manually reading thousands of responses to find trends.
Auto-tag images, video, and product descriptions into your catalog taxonomy. Scale to millions of SKUs or assets without a manual QA pass.
Start free. Scale as you grow. No hidden fees.
Each call to POST /v1/classify counts as one request, regardless of input size or the number of categories in your schema.
Both options work. You can pass a categories array directly in each request, reference a saved schema_id, or omit both to get generic AI-generated labels.
JPEG, PNG, GIF, and WebP. You can send a public URL or a base64-encoded string. Max file size is 20MB per image.
Most requests complete synchronously in under 500ms. For large media files, the API returns a job_id you can poll, or configure a webhook to receive the result automatically.
Requests above your limit return a 429 with a clear error message. Upgrade your plan anytime - limits reset on the 1st of each month.
Yes - any automation tool that can make an HTTP request works out of the box. Add an HTTP node, set your Authorization header, pass your input and categories as JSON, and branch your workflow on the returned label. No SDK required.
Absolutely. This is one of the most common use cases. Pass the form submission text and categories like ["genuine_inquiry","spam","abuse"] - you'll get a label and confidence score back in under 300ms, fast enough to block or flag the submission before your server processes it.
No. Input data submitted to the API is never used to train models. See our Privacy Policy for full details.
Pick a time and we'll walk you through the API, answer integration questions, or discuss an Enterprise plan. Spots are limited - book directly below.
Choose a day for your session (Eastern Time)
Free plan, no credit card required. One API call away from smarter routing, spam blocking, and automated triage.
Start for free