Excited to launch a multi-year partnership bringing Fireworks to Microsoft Azure Foundry! Learn more

Fireworks AI on Microsoft Foundry

State-of-the-Art Open Model Inference. Now Native on Azure Foundry.

Fireworks AI is now available as a first-party inference provider inside Microsoft Foundry, enabling developers to deploy and scale open-weight models with enterprise-grade infrastructure

Start Building

A Unified Platform for Open Models

Across industries, organizations are standardizing on open models to gain greater control over performance, cost, customization, and compliance


With Fireworks AI on Microsoft Foundry, teams can:

  • •Deploy state-of-the-art open models
  • •Bring their own weights
  • •Optimize inference performance
  • •Scale workloads across Azure infrastructure

All from one unified platform

Start Building with Fireworks AI Today

Deploy state-of-the-art open models directly inside Microsoft Foundry and scale across Azure infrastructure

Get Started Today

Best-in-Class Open Model Performance

Fireworks AI provides a high-performance inference engine built specifically for modern open-weight models


Now available natively within Microsoft Foundry, this engine enables:

  • •High-throughput inference
  • •Ultra-low latency responses
  • •Cost-efficient token generation
  • •Production-grade reliability on Microsoft Azure

Developers get the performance needed to run large-scale AI applications without managing infrastructure

Developer Flexibility: Bring Your Own Weights

Organizations can upload and serve their own fine-tuned or quantized models

Capabilities include:

  • •Upload custom model weights
  • •Register models directly in Foundry
  • •Serve with Fireworks inference optimization
  • •Deploy without modifying the serving stack

This allows teams to combine open-model innovation with enterprise-grade deployment