Semantic search as a service
The semantic search API for your catalogue
Upload your products, articles or FAQs, train the list in one click and add a search to your site or app that understands meaning and tolerates typos. No AI infrastructure to run.
€5 welcome credit · No card required · Free playground
cordless earphones for running
- Bluetooth sports headphones0.94
- Wireless noise-cancelling headphones0.88
- Bluetooth headset with microphone0.79
How it works
From your data to a search engine in three steps
No need to know about embeddings or deploy models: XEYE handles it.
Step 1
Create a list
Group what you want to be searchable: products, help articles, FAQs… Add elements by hand or import a JSON file in one go.
Step 2
Train it
In one click XEYE computes the embeddings for the list and, if you want, enriches each element with AI-generated descriptions. You see the price before launching.
Step 3
Search from your app
Create an API key and call a single HTTP endpoint. You get the results ranked by relevance, each with your own data attached.
Hybrid search
It understands what your users mean
Every search combines two signals and returns a single score per result.
Fuzzy matching
Tolerates typos, partial words and variations: “runing shoos” finds “running shoes”.
Semantic matching
Compares meanings, not letters: “cordless earphones” finds “Bluetooth headphones” even though they share no words.
Transparent scoring
Each result carries the combined score and, on request, its text and semantic breakdown.
“cordless earphones for running”
Bluetooth sports headphones
- Combined score
- 0.94
- Text
- 0.31
- Semantic
- 0.97
“womens runing shoos”
Women's running shoes
- Combined score
- 0.96
- Text
- 0.90
- Semantic
- 0.93
API integration
Integrate it in minutes
One endpoint, one API key and JSON. It works with any language or framework that can make HTTP requests.
- Authenticate with your API key in the X-API-Key header.
- Each result returns the parameters you stored (SKU, URL, price…): no second lookup needed.
- Per-account rate limit with standard headers, and errors with a machine-readable code.
- Per-list search log to see what your users look for.
Request
curl -X POST https://search.xeye.es/api/v1/search \
-H "Content-Type: application/json" \
-H "X-API-Key: $XEYE_API_KEY" \
-d '{
"list_name": "products",
"search_term": "wireless headphones",
"limit": 5
}'Response
{
"success": true,
"results": [
{
"item": "Sony WH-1000XM5 wireless headphones",
"score": 0.93,
"params": {
"sku": "SONY-XM5",
"url": "/products/sony-wh-1000xm5",
"price": 348
}
}
],
"total_results": 1,
"duration_ms": 42
}Features
Everything you need for a production search
Hybrid search
Semantic and fuzzy in a single query, with one combined score per result.
AI-generated descriptions
Optionally, a language model expands each element with context so it can be found in more ways. You can turn it off per list.
Your data in every result
Store free-form JSON parameters with each element (SKU, URL, price…) and get them back in the response.
Secure API keys
Stored hashed and shown only once. Create one per environment and revoke them instantly.
Free playground
Try searches on any of your lists from the console, without a key and without spending credit.
Search log
See what is searched in each list and spot what your users cannot find.
Use cases
What it is used for
Online shops
A product search that gets it right even when the customer does not use the exact words in the catalogue.
Help centres
Take each user to the article or FAQ that answers their question, asked in their own words.
Directories and internal catalogues
Search across references, suppliers, documents or records without relying on exact codes.
Assistants and chatbots
Retrieve the most relevant entry of your knowledge base to hand to your assistant.
Pricing
Pay only for what you use
Prepaid credit, no monthly fees and no lock-in.
- Trainings: you see the price before launching and, if one fails, it is refunded in full.
- API searches: a small amount per search served.
- Console playground: free, no credit spent.
- Email alerts when your balance drops below the threshold you choose.
Frequently asked questions
Frequently asked questions about XEYE
What is XEYE?
XEYE is a semantic search platform delivered as a service. You upload a list of elements (products, articles, FAQs…), train it in one click and query it from your website or application through an HTTP API with an API key.
What is semantic search?
It is a search that compares the meaning of the query with the meaning of each element, instead of comparing words only. That is why “cordless earphones” finds “Bluetooth headphones”. XEYE combines it with fuzzy text matching, which tolerates typos and partial words.
How do I integrate XEYE into my website or application?
You create an API key in the console and send a POST request to /api/v1/search with the list name and the search text. The response is JSON with the results ranked by relevance and the parameters of each element. It works from any language that can make HTTP requests.
Do I need AI knowledge or my own infrastructure?
No. XEYE computes the embeddings, hosts the index and serves the searches. You only manage your lists and call the API.
How much does XEYE cost?
XEYE runs on prepaid credit, with no monthly fees. You get €5 of credit when you sign up. Trainings and searches served by the API spend credit; the console playground is free.
What data can I make searchable?
Any collection of texts: products, articles, frequently asked questions, records, references… Each element has a text, an optional description and free-form JSON parameters that are returned with every result.
Is my data private?
You decide which lists are public (queryable with your API key) and which are private (console only). API keys are stored hashed and AI enrichment can be turned off per list.
Start searching by meaning today
Create your account, upload a list and run your first search in about five minutes.