Instantly deploy ML production inference to thousands of dedicated GPUs - at the lowest cost in the market.

  • What is SIE?
    • SIE is a one-click API that allow you to scale AI/ML inferences to infinity without configuring infrastructure, all at the lowest cost in the market.
    • SIE API is the lowest priced API today, powered by Salad’s unique distributed architecture.
  • How does SIE work?
    • Salad Inference Endpoints (SIE) allows developers to deploy deep learning models for production-scale inference via simple API calls. Teams accustomed to hyperscale solutions will find the workflow similar to that of AWS Sagemaker Inference APIs.

Inference Endpoints Marketplace

Navigate to the Inference Endpoints Marketplace by clicking the Inference Endpoints link in the left sidebar. You will see the models that are currently available.

API Endpoints


To use the Inference Endpoint APIs, you will need your Salad API Key. You can find the API Key in the upper right hand Account drop down menu.

Need file storage?

  • You can use a file hosting service like Google Drive, Dropbox, or a public S3 bucket.
  • or try the new S4 - Salad Simple Storage Service
    • Starting immediately, salad users can upload and download files to storage-api.salad.com, authenticating with either their salad api key, or the jwt issued by the instance metadata service. These files will be removed automatically after 30 days, so customers should not use this for permanent storage. There is currently a maximum file size of 100mb.


  • Send a POST request to the API Endpoint with your file (if required)
    • The response will return a job id
  • Send a GET request with that job id at the end of the URL
    • The response will return your desired output
  • Example: Using the Whisper Large v3 API for Transcription results via Postman
    • Open postman and create a new HTTP request
    • Set the HTTP method to POST
    • Enter the API endpoint URL
    • Add API Key Authorization
      • Key: Salad-Api-Key
      • Value: provide your Salad API Key
      • Add to: header
    • Add the Request Body
      • select the raw option, and choose JSON from the dropdown menu
      • Add the URL of your audio file in the request body as shown in the example.
        • the file URL must be a direct download link
    • Send the Request
      • Click the Send button. The API will respond with the job id
    • Send another GET request with that job id at the end of the URL
      • Click the Send button. The API will respond with the transcription result in JSON format.


  • SIE is being offered for free for a limited time!