canonical: https://jentic.com/apis/genderize.io/genderize

# Genderize API

The Genderize API predicts the most likely gender associated with a first name, returning a gender label, a probability score between 0 and 1, and the count of source records used to derive the prediction. It supports localisation by ISO country and language code, batch lookups of up to 10 names per request, and a free tier of 100 requests per day. The single endpoint is a pure read; an optional API key in the apikey query parameter unlocks higher quotas.

## For AI agents

Predict the likely gender of a first name with a confidence score, with optional country and language localisation.

## Scope

Does not handle full identity verification, age prediction, or nationality lookup - use for first-name gender prediction only.

## Capabilities

- Predict the most likely gender for a single first name with /?name={name}
- Batch up to 10 names in one call by repeating the name parameter (name[]=)
- Localise predictions by ISO 3166-1 country code with the country_id query parameter
- Localise predictions by ISO 639-1 language code with the language_id query parameter
- Read remaining-quota headers from each response to track usage against the 100/day free tier

## Use cases

### CRM Contact Enrichment

Enrich a CRM contact list with predicted gender and probability so segmentation and personalisation logic can run downstream. The agent batches names 10 at a time against the root endpoint and writes the gender, probability, and count fields back to the contact record. Removes a manual or vendor-paid pre-processing step.

Example prompt: Build a request with name[]=Alice&name[]=Bob (up to 10), GET /, and write each {name, gender, probability} pair back to the CRM record

### Localised Gender Predictions For International Audiences

Improve prediction accuracy for international name lists by passing country_id or language_id with each call. The agent groups records by region and issues localised lookups so 'Andrea' resolves correctly in Italy versus the US. Sharpens downstream analytics on multi-region datasets.

Example prompt: GET /?name=Andrea&country_id=IT for the Italian rows and /?name=Andrea&country_id=US for the US rows; merge results back onto the source dataset

### Form Validation And Personalisation

When a user signs up with only a first name, predict gender to drive default avatar selection or pronoun fallback while leaving an explicit override option. The agent calls the endpoint at signup and stores the result with its probability so the UI can fall back to a neutral default when confidence is low.

Example prompt: On signup, GET /?name={firstName}; if probability < 0.7, render a neutral avatar and prompt for explicit pronouns

### Agent-Driven Enrichment Through Jentic

AI agents discover Genderize via Jentic intent search, load the predict-gender operation, and enrich names inline without writing a custom client. Jentic isolates the API key in its vault for the higher-quota tier. Useful for data-prep agents working over CRM exports.

Example prompt: Use Jentic search query 'predict the gender of a name' to load the genderize_predict_gender tool and execute with the name and optional country_id

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| GET | / | Predict gender for a single name or a batch with optional localisation |

## Key resources

- **Gender Prediction** — Predict likely gender for a single name or a batch with confidence scoring

## Why Jentic

- **Setup:** Wiring Genderize by hand means registering your apikey, appending it to the query string on every call, and handling the response parsing yourself. Through Jentic you install once, import Genderize from the API Directory, store the key once, and your agent calls it.
- **Permission scoping:** Genderize exposes a single name-prediction operation that takes its input in the query string, so scope the agent to just the operation it needs. You choose the operations it may call, so nothing beyond gender prediction runs unless you add it.
- **Credential handling:** Your Genderize apikey is stored once, encrypted, by your own Jentic One instance and appended to the request at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'predict the gender of a first name', and Jentic returns the Genderize operation with its input schema, including the optional country_id and language_id parameters, so the agent calls it directly without browsing the reference docs.

## Related APIs

- **Agify API** — Agify predicts age from a first name from the same family of single-purpose enrichment APIs.
- **Nationalize API** — Nationalize predicts likely nationality from a first name and shares Genderize's interface conventions.
- **Abstract API** — Abstract API offers a broader bundle of enrichment endpoints; Genderize stays a single-purpose, low-cost gender prediction.

## FAQ

### What authentication does the Genderize API use?

Genderize is keyless on the free tier (100 requests per day per IP). For higher quotas pass an apikey query parameter on each request. Through Jentic the key is stored encrypted in the vault and appended at execution time.

### Can I batch multiple names in one Genderize request?

Yes. The single GET / endpoint accepts up to 10 names per call by repeating the name parameter as name[]=Alice&name[]=Bob. The response is an array of predictions in the same order as the inputs.

### How do I localise a prediction by country?

Pass country_id with an ISO 3166-1 alpha-2 code, for example country_id=IT. Genderize will weight its prediction by data from that country, sharpening results for ambiguous cross-cultural names.

### What are the rate limits for the Genderize API?

The free tier allows 100 requests per day per IP. Paid plans raise the daily ceiling - Genderize sets X-Rate-Limit headers on each response so clients can track remaining quota in real time.

### What does the probability field mean?

Probability is a value between 0 and 1 representing the share of source records that match the predicted gender. Treat values below roughly 0.7 as low-confidence and either store the score or fall back to a neutral default in your UI.

### How do I predict gender through Jentic?

Install with pip install jentic, search 'predict the gender of a name', load the schema for the genderize_predict_gender operation, and execute with the name and optional country_id. Jentic appends the API key from the vault.

### Can I limit what my agent is allowed to do with the Genderize API?

Yes. Genderize exposes a single name-prediction operation that takes its input in the query string, and because Jentic One is self-hosted your own rules decide which operations and credentials the agent may use. You scope the agent to just that gender-prediction call, so nothing beyond it runs unless you explicitly add another operation. Your apikey is stored once by your own instance and appended at execution time rather than exposed to the agent.