The Samsara APIs have rate limits in place to protect our services from API traffic spikes that could put our servers at risk. If you exceed any of the rate limits described below, Samsara will respond with a 429 error code.

📘

Ad-Hoc Rate Limit Changes

Samsara reserves the right to change rate limits for system protection. As a principle, Samsara will plan to give a heads up before any changes are made. Changes do not happen often.

Global Rate Limits

These rate limit ceilings that apply to all API endpoints. Some API endpoints have more restrictive rate limits. See Per API Rate Limits below for more details.

• Per token
  • Each API access token may make 150 API requests per second
• Per organization
  • Each organization may make 200 API requests per second

In other words, a single API token may only make 150 requests per second. An organization may have multiple API tokens, but an organization may only make 200 requests per second total.

Endpoint-level Rate Limits

These rate limits apply to specific endpoints within an organization. Read the description for each individual endpoint in the API Reference to see if it has an endpoint-specific rate limit. See the equipment stats endpoint as an example. Not all endpoints have rate limits documented in the API reference, so refer to the below chart for more.

Endpoint-level Rate Limit Categories

There are 5 different rate limit categories. Each endpoint falls into one of these tiers. These rate limits are applied on a per organization basis.

List of Rate Limits

🚧

The following Rate Limits will take effect starting October 23rd, 2023

The below are new endpoint-level rate limits. They will be rolling out starting October 23rd, 2023. For more information on how this rollout will be conducted and how it may impact you, refer to the Changelog post here. For further assistance, reach out to Samsara Support.

Handling Rate Limits

When an application hits a rate limit, the API will return an API response with a HTTP status code of 429 Too Many Requests.

The response will include the following header:

You should use the Retry-After header in order to appropriately back-off the endpoint so that you do not get rate-limited again.

Below is a sample Python script that retrieves the current snapshot of the locations of all of your vehicles data from our API while gracefully handling rate limits. The script is designed to adhere to the API’s rate-limiting guidelines by utilizing the Retry-After header. This ensures that the script will pause and retry the request based on the guidance, thus minimizing the chances of encountering a “429 Too Many Requests” error. Feel free to modify it to suit your specific endpoints and convenience.

import requests  
import time

def make_api_request(url, headers):

    while True:  #Loop for retries
        response = requests.get(url, headers=headers)
        
        if response.status_code != 429:  #Status other than 'Too Many Requests'
            return response
        
        #Encountered rate limit, get the time to wait from the 'Retry-After' header
        retry_after = float(response.headers.get('Retry-After', 1))  #Default to 1 second if header is missing
        
        print(f"Rate limit exceeded. Retrying in {retry_after} seconds.")
        
        # Wait for the duration specified in 'Retry-After' before making another request
        time.sleep(retry_after)

if __name__ == '__main__':
    url = "https://api.samsara.com/fleet/vehicles/stats?types=gps"  #Replace with the actual API endpoint
    headers = {"Authorization": "Bearer YOUR_API_TOKEN"}  #Replace with your actual access token

    response = make_api_request(url, headers)
    
    if response:
        print(f"Successfully received data: {response.json()}")
    else:
        print("Failed to retrieve data.")