Guide · Python · Validation

LEI Validation, Search & LOU Lookup in Python

Three API endpoints for Legal Entity Identifiers — validate with MOD-97 checksum, search 2.3 million entities by name, and list all Local Operating Units. Here's how it all fits together using Python and the requests library.

In this guide

1. What is a LEI?
2. LEI structure — the 20-character code
3. The MOD-97 checksum
4. Three endpoints — validate, search, LOUs
5. Endpoint 1: Validate a single LEI
6. Endpoint 2: Search entities by name
7. Endpoint 3: List all LOUs
8. Python code examples
9. cURL examples
10. Use cases — KYC, MiFID II, EMIR

1. What is a LEI?

A Legal Entity Identifier (LEI) is a 20-character alphanumeric code that uniquely identifies a legal entity participating in financial transactions. The standard is ISO 17442, managed by the Global Legal Entity Identifier Foundation (GLEIF).

LEIs are mandatory for:

Derivatives reporting under EMIR and Dodd-Frank
MiFID II transaction reporting in the EU
Securities financing transactions (SFTR)
Any entity trading on regulated markets

The GLEIF database contains over 2.3 million active LEIs, covering entities in 250+ jurisdictions.

Entity	LEI
Goldman Sachs	784F5XWPLTWKTBV3E584
Deutsche Bank	7LTWFZYICNSX8D621K86
Apple Inc	HWUPKR0MPOU8FGXBT394

ℹ️Unlike VAT numbers or REGON, LEI is a global standard — the same 20-character format works across all jurisdictions. A German bank and a Japanese trading firm use the same LEI structure.

2. LEI structure — the 20-character code

A LEI is exactly 20 characters: a 4-character LOU prefix + a 14-character entity identifier + 2 check digits.

Part	Length	Description	Example (Goldman Sachs)
LOU prefix	4	Identifies the Local Operating Unit that issued the LEI	784F
Entity	14	Unique entity identifier (alphanumeric, no vowels in prefix)	5XWPLTWKTBV3E5
Check digits	2	MOD-97 checksum (identical to IBAN)	84

All characters are uppercase alphanumeric (A-Z, 0-9). The LOU prefix is assigned by GLEIF to each Local Operating Unit — there are about 40 active LOUs worldwide.

3. The MOD-97 checksum

The LEI checksum uses the same MOD-97 algorithm as IBAN (ISO 7064):

Take the full 20-character LEI

Convert each letter to a number: A=10, B=11, ..., Z=35

Concatenate all digits into a single large number string

Calculate the remainder when dividing by 97

If the remainder equals 1, the LEI is valid

Worked example: 784F5XWPLTWKTBV3E584

Each character is converted to its numeric equivalent. Digits stay as-is; letters become two-digit numbers (A=10, B=11, ..., Z=35):

784F → 7, 8, 4, 15 (F=15)
5XWPLTWKTBV3E5 → 5, 33, 32, 25, 21, 29, 32, 20, 29, 11, 31, 3, 14, 5
84 → 8, 4

The full digit string is: 7841553325212932202911313141584. Computing this number modulo 97 yields 1 — so the LEI is valid.

Python implementation

def validate_lei_checksum(lei: str) -> bool:
    """Validate a LEI using the ISO 17442 MOD-97 checksum.

    Args:
        lei: A 20-character alphanumeric LEI string.

    Returns:
        True if the checksum is valid, False otherwise.
    """
    digits = ""
    for ch in lei:
        if ch.isalpha():
            digits += str(ord(ch) - 55)  # A=10, B=11, ..., Z=35
        else:
            digits += ch

    # Python handles arbitrarily large integers — no chunking needed
    return int(digits) % 97 == 1


print(validate_lei_checksum("784F5XWPLTWKTBV3E584"))  # True
print(validate_lei_checksum("784F5XWPLTWKTBV3E500"))  # False

⚠️A valid checksum does not mean the entity exists or is currently active. LEIs expire annually — an entity that did not renew its LEI will still pass the checksum but have a LAPSED registration status. Always check registrationStatus from the API response.

4. Three endpoints — validate, search, LOUs

The IsValid LEI API provides three endpoints that cover the full spectrum of LEI operations — from validating a single code to searching across 2.3 million entities:

Validate a single LEI

GET /v0/lei?value=... — MOD-97 checksum + entity lookup from the GLEIF database (2.3M records) with fallback to the GLEIF REST API. Returns full entity details including legal name, country, status, and the issuing LOU.

Search entities by name

GET /v0/lei/search?q=... — Fuzzy search across 2.3M legal entity names using PostgreSQL trigram similarity. Filter by country and entity status. Paginated results sorted by relevance.

List all LOUs

GET /v0/lei/lous — Complete list of ~40 Local Operating Units with their LEI, name, country, and status. Useful for building dropdowns, validating LOU prefixes, or understanding who issued a given LEI.

Get your free API key at isvalid.dev. The free tier includes 100 calls per day with no credit card required.

Full parameter reference and response schema: LEI Validation API docs →

5. Endpoint 1: Validate a single LEI

The validate endpoint performs a multi-step verification chain:

Format check

20 alphanumeric characters (A-Z, 0-9)

MOD-97 checksum

ISO 7064 checksum — remainder must equal 1

Local database lookup

GLEIF Golden Copy (~2.3M records) — dataSource: "gleif-db"

GLEIF API fallback

If not found locally, query GLEIF REST API — dataSource: "gleif-api"

LOU resolution

LOU info resolved in parallel from the lei_lous table

Response — entity found (Goldman Sachs)

{
  "valid": true,
  "lei": "784F5XWPLTWKTBV3E584",
  "louCode": "784F",
  "checkDigits": "84",
  "found": true,
  "dataSource": "gleif-db",
  "entity": {
    "legalName": "GOLDMAN SACHS GROUP INC",
    "country": "US",
    "entityStatus": "ACTIVE",
    "registrationStatus": "ISSUED",
    "category": null,
    "initialRegistrationDate": "2012-06-06",
    "lastUpdate": "2024-08-14",
    "nextRenewal": "2025-08-14",
    "managingLou": "EVK05KS7XY1DEII3R011"
  },
  "lou": {
    "lei": "EVK05KS7XY1DEII3R011",
    "name": "Bloomberg Finance L.P.",
    "country": "US",
    "status": "ACTIVE"
  }
}

Response — invalid checksum

{
  "valid": false
}

Response — valid checksum but entity not found

{
  "valid": true,
  "lei": "784F5XWPLTWKTBV3E584",
  "louCode": "784F",
  "checkDigits": "84",
  "found": false,
  "dataSource": "gleif-api",
  "entity": null,
  "lou": null
}

Field	Type	Description
valid	boolean	Whether the LEI passed format and MOD-97 checksum validation
lei	string	The normalised 20-character LEI (only present when `valid: true`)
louCode	string	First 4 characters — the LOU prefix (only present when `valid: true`)
checkDigits	string	Last 2 characters — the MOD-97 check digits (only present when `valid: true`)
found	boolean	Whether the entity was found in the GLEIF database or API (only present when `valid: true`)
dataSource	string \| null	`"gleif-db"` (local database) or `"gleif-api"` (GLEIF REST API fallback)
entity	object \| null	Entity details — legal name, country, status, registration dates, managing LOU
lou	object \| null	LOU details — LEI, name, country, and status of the issuing Local Operating Unit

6. Endpoint 2: Search entities by name

The search endpoint lets you find legal entities by name using fuzzy matching powered by PostgreSQL's pg_trgm extension. It supports filtering by country and entity status, with paginated results sorted by similarity score.

Parameters

Parameter	Required	Description
q	Yes	Search query, minimum 2 characters
country	No	ISO 3166-1 alpha-2 code, e.g. `"US"`, `"DE"`, `"PL"`
entityStatus	No	`"ACTIVE"` or `"INACTIVE"`
page	No	Page number, default 1
limit	No	Results per page, default 20, max 100

Response

{
  "results": [
    {
      "lei": "784F5XWPLTWKTBV3E584",
      "legalName": "GOLDMAN SACHS GROUP INC",
      "country": "US",
      "entityStatus": "ACTIVE",
      "registrationStatus": "ISSUED",
      "category": null
    }
  ],
  "page": 1,
  "limit": 20,
  "total": 47
}

✅The search is fuzzy — ?q=goldmn sachs will still find Goldman Sachs. Results are sorted by similarity score, so the best matches come first.

7. Endpoint 3: List all LOUs

No parameters needed. Returns all ~40 Local Operating Units — the organizations authorized by GLEIF to issue and maintain LEIs. Each LOU has a 4-character code that appears as the prefix of every LEI it issues.

Response

{
  "lous": [
    {
      "louCode": "784F",
      "lei": "EVK05KS7XY1DEII3R011",
      "name": "Bloomberg Finance L.P.",
      "country": "US",
      "status": "ACTIVE"
    }
  ],
  "total": 40
}

LOUs are responsible for the initial registration and annual renewal of LEIs. When a LEI lapses (the entity did not renew), it is the managing LOU that marks the registration as LAPSED. The LOUs endpoint is useful for building filter dropdowns, validating LOU prefixes in bulk, or understanding the geographic distribution of LEI issuance — most LOUs are concentrated in Europe and North America, but they serve entities globally.

8. Python code examples

Using the requests library — the de facto standard for HTTP in Python. Install it with pip install requests. All three endpoints share the same authentication pattern.

Example 1: Validate a single LEI

# lei_client.py
import os
import requests

API_KEY = os.environ["ISVALID_API_KEY"]
BASE_URL = "https://api.isvalid.dev"


def validate_lei(lei: str) -> dict:
    """Validate a LEI using the IsValid API.

    Args:
        lei: A 20-character LEI string.

    Returns:
        Validation result as a dictionary.

    Raises:
        requests.HTTPError: If the API returns a non-2xx status.
    """
    response = requests.get(
        f"{BASE_URL}/v0/lei",
        params={"value": lei},
        headers={"Authorization": f"Bearer {API_KEY}"},
    )
    response.raise_for_status()
    return response.json()


# ── Example usage ────────────────────────────────────────────────────────────

result = validate_lei("784F5XWPLTWKTBV3E584")

if not result["valid"]:
    print("Invalid LEI")
else:
    print(f"LEI     : {result['lei']}")
    print(f"LOU     : {result['louCode']}")
    print(f"Found   : {result['found']}")
    print(f"Source  : {result['dataSource']}")
    if result.get("entity"):
        print(f"Name    : {result['entity']['legalName']}")
        print(f"Country : {result['entity']['country']}")
        print(f"Status  : {result['entity']['entityStatus']}")

Example 2: Search for entities

def search_lei(
    query: str,
    *,
    country: str | None = None,
    entity_status: str | None = None,
    page: int = 1,
    limit: int = 20,
) -> dict:
    """Search for legal entities by name.

    Args:
        query: Search string (min 2 characters).
        country: ISO 3166-1 alpha-2 filter (e.g. "US", "DE").
        entity_status: "ACTIVE" or "INACTIVE".
        page: Page number (default 1).
        limit: Results per page (default 20, max 100).

    Returns:
        Search results with pagination metadata.
    """
    params: dict = {"q": query, "page": page, "limit": limit}
    if country:
        params["country"] = country
    if entity_status:
        params["entityStatus"] = entity_status

    response = requests.get(
        f"{BASE_URL}/v0/lei/search",
        params=params,
        headers={"Authorization": f"Bearer {API_KEY}"},
    )
    response.raise_for_status()
    return response.json()


# Search for German banks
banks = search_lei("deutsche bank", country="DE", entity_status="ACTIVE")

print(f"Found {banks['total']} results (page {banks['page']})")
for r in banks["results"]:
    print(f"  {r['lei']} — {r['legalName']} ({r['country']})")

Example 3: List LOUs

def list_lous() -> dict:
    """Fetch all Local Operating Units.

    Returns:
        Dictionary with "lous" list and "total" count.
    """
    response = requests.get(
        f"{BASE_URL}/v0/lei/lous",
        headers={"Authorization": f"Bearer {API_KEY}"},
    )
    response.raise_for_status()
    return response.json()


data = list_lous()
print(f"{data['total']} LOUs worldwide:")
for lou in data["lous"]:
    print(f"  {lou['louCode']} — {lou['name']} ({lou['country']})")

Full example: Flask integration

In a KYC onboarding form, you might use it like this with Flask:

# app.py (Flask)
from flask import Flask, request, jsonify

app = Flask(__name__)


@app.post("/onboard")
def onboard():
    data = request.get_json()

    try:
        lei_check = validate_lei(data["lei"])
    except requests.RequestException:
        return jsonify(error="LEI validation service unavailable"), 502

    if not lei_check["valid"]:
        return jsonify(error="Invalid LEI format"), 400

    if not lei_check["found"]:
        return jsonify(error="LEI not found in GLEIF database"), 400

    entity = lei_check["entity"]
    if entity["registrationStatus"] == "LAPSED":
        return jsonify(error="LEI registration has lapsed — please renew"), 400

    # Entity is valid, found, and current — proceed with onboarding
    return jsonify(
        success=True,
        entityName=entity["legalName"],
        country=entity["country"],
        lou=lei_check["lou"]["name"] if lei_check.get("lou") else None,
    )

✅The API strips whitespace and converts to uppercase automatically. Pass raw user input directly — you do not need to pre-process it.

9. cURL examples

Validate a LEI (Goldman Sachs):

curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.isvalid.dev/v0/lei?value=784F5XWPLTWKTBV3E584"

Search by name (Goldman Sachs, US only):

curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.isvalid.dev/v0/lei/search?q=goldman%20sachs&country=US"

Search with pagination (page 2, 5 results per page):

curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.isvalid.dev/v0/lei/search?q=bank&limit=5&page=2"

List all LOUs:

curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.isvalid.dev/v0/lei/lous"

10. Use cases — KYC, MiFID II, EMIR

LEI validation is a building block in many regulatory and compliance workflows. Here are the most common scenarios:

KYC & Onboarding

Verify counterparty identity during customer onboarding. Validate the LEI, check entity status (ACTIVE vs INACTIVE), and confirm the registration is not LAPSED. Combine with IBAN and VAT validation for a complete KYC check.

MiFID II Transaction Reporting

EU investment firms must include LEIs in transaction reports. The validate endpoint confirms both format and entity existence. Use the search endpoint to help users find their LEI when they don't know it — fuzzy matching makes this reliable even with partial or misspelled names.

EMIR Derivatives Reporting

Both counterparties to an OTC derivative trade must be identified by LEI. Validate LEIs before submission to trade repositories. LAPSED registrations are rejected by regulators — always check registrationStatus before accepting a LEI for reporting purposes.

Summary

❌Do not rely on checksum alone — always check found and registrationStatus from the API

❌Do not assume a valid LEI is currently active — LEIs expire annually and can lapse

❌Do not store LEIs as lowercase — the format is uppercase alphanumeric

✅Use the validate endpoint for single-LEI checks with full entity details

✅Use the search endpoint to help users find LEIs by company name

✅Check registrationStatus — only "ISSUED" means the LEI is current and renewed

Try LEI validation instantly

Free tier includes 100 API calls per day. No credit card required. Validate, search, and look up LOUs — all included.

Get Free API Key Browse API Docs

LEI Validation, Search & LOU Lookup in Python

1. What is a LEI?

2. LEI structure — the 20-character code

3. The MOD-97 checksum

Worked example: 784F5XWPLTWKTBV3E584

Python implementation

4. Three endpoints — validate, search, LOUs

Validate a single LEI

Search entities by name

List all LOUs

5. Endpoint 1: Validate a single LEI

Response — entity found (Goldman Sachs)

Response — invalid checksum

Response — valid checksum but entity not found

6. Endpoint 2: Search entities by name

Parameters

Response

7. Endpoint 3: List all LOUs

Response

8. Python code examples

Example 1: Validate a single LEI

Example 2: Search for entities

Example 3: List LOUs

Full example: Flask integration

9. cURL examples

10. Use cases — KYC, MiFID II, EMIR

KYC & Onboarding

MiFID II Transaction Reporting

EMIR Derivatives Reporting

Summary

See also

Try LEI validation instantly