Product

Warehouse every request from OpenAI

Velvet is an AI gateway. Our proxy warehouses every OpenAI request to your PostgreSQL database. Swap two lines of code to get started.

Use Velvet with Anthropic

Velvet's AI gateway logs each response from the OpenAI API to your database. This gives you queryable data to observe, analyze, and optimize your AI features. Velvet is made to be composable with your unique implementation. Common use cases include:

  • Cost analysis between endpoints or models
  • Response observability and tracing
  • Caching, batching, and dataset generation

Set Velvet as a proxy to warehouse OpenAI requests to PostgreSQL. We support request warehousing for most endpoints, and act as a transparent proxy for anything else.

  • /v1/chat/completions
  • /v1/batches
  • v1/files
  • /v1/moderations
  • /v1/embeddings

As a result - you'll have additional data to leverage, and you'll never lose API functionality.

How to get started

Simply swap two lines of code to start warehousing every request.

Learn more about Velvet and OpenAI to get started.

Email team@usevelvet.com or schedule a call with any questions.

AI gateway

Analyze, evaluate, and monitor your AI

Free up to 10k requests per month.

2 lines of code to get started.

Try Velvet for free

More articles

Product
Run model experiments on your historical data

Test models, settings, and metrics against historical request logs.

Product
Query logs with Velvet's text-to-SQL editor

Use our data copilot to query your AI request logs with SQL.

Product
Warehouse every request from OpenAI

Use Velvet to observe, analyze, and optimize your AI features.