Product

Developers

Solutions

Resources

Product

Developers

Solutions

Resources

AI Gateway

AI Gateway

AI Gateway

Fast, Reliable AI Gateway

Fast, Reliable AI Gateway

Fast, Reliable AI Gateway

Connect, load balance, setup fallbacks and seamlessly manage

200+ AI models using an OpenAI compatible API.

Connect, load balance, setup fallbacks and seamlessly manage

200+ AI models using an OpenAI compatible API.

Connect, load balance, setup fallbacks and seamlessly manage

200+ AI models using an OpenAI compatible API.

npx @portkey-ai/gateway
npx @portkey-ai/gateway

🫡 The AI Gateway has served over 2 billion requests for teams at

Unified Interface

Unified Interface

Connect to 250+ LLM providers through a unified API. Simplify interactions and minimize integration hassles.

Connect to 250+ LLM providers through a unified API. Simplify interactions and minimize integration hassles.

Load Balancing

Load Balancing

Ensure optimal performance by distributing workloads across multiple keys and providers.

Ensure optimal performance by distributing workloads across multiple keys and providers.

Fallbacks

Fallbacks

Automatically switch LLMs during failures to deliver uninterrupted & accurate performance.

Automatically switch LLMs during failures to deliver uninterrupted & accurate performance.

Semantic Cache

Semantic Cache

Speed up repeated queries, enhance response times, and reduce costs.

Speed up repeated queries, enhance response times, and reduce costs.

Multimodality

Multimodality

The Gateway integrates with multimodal models through the same unified API and supports vision, audio, image generation, and more capabilities across providers.

The Gateway integrates with multimodal models through the same unified API and supports vision, audio, image generation, and more capabilities across providers.

Enterprise Ready

Portkey partners with F5 (the creators of Nginx) to deploy the AI Gateway to F5's Global Distributed Cloud

Portkey partners with F5 (the creators of Nginx) to deploy the AI Gateway to F5's Global Distributed Cloud

Ready to try Portkey?

By signing up, you agree to the privacy policy & terms of service

Ready to try Portkey?

By signing up, you agree to the privacy policy & terms of service

Ready to try Portkey?

By signing up, you agree to the privacy policy & terms of service

Next Feature

Prompt Management

FAQ

Got questions?

If you have any other questions - please get in touch at [email protected]

What types of models can I use with AI Gateway?

AI Gateway supports 250+LLMs from leading platforms like OpenAI, Anthropic, Google Gemini, Meta, Mistral, Groq, and many more, including your locally or privately hosted models on Ollama, On-prem, or more.

What types of models can I use with AI Gateway?

AI Gateway supports 250+LLMs from leading platforms like OpenAI, Anthropic, Google Gemini, Meta, Mistral, Groq, and many more, including your locally or privately hosted models on Ollama, On-prem, or more.

What types of models can I use with AI Gateway?

AI Gateway supports 250+LLMs from leading platforms like OpenAI, Anthropic, Google Gemini, Meta, Mistral, Groq, and many more, including your locally or privately hosted models on Ollama, On-prem, or more.

Does the AI Gateway change the behavior of the LLM?

The AI Gateway by design does not alter or affect the behavior of the underlying LLM in any way. It is a service running on top of your LLM API calls to make them interoperable, reliable, and robust. That said, with the Guardrails on the Gateway framework, users can run real-time guardrails on user inputs and LLM outputs to validate and route requests appropriately.

Does the AI Gateway change the behavior of the LLM?

The AI Gateway by design does not alter or affect the behavior of the underlying LLM in any way. It is a service running on top of your LLM API calls to make them interoperable, reliable, and robust. That said, with the Guardrails on the Gateway framework, users can run real-time guardrails on user inputs and LLM outputs to validate and route requests appropriately.

Does the AI Gateway change the behavior of the LLM?

The AI Gateway by design does not alter or affect the behavior of the underlying LLM in any way. It is a service running on top of your LLM API calls to make them interoperable, reliable, and robust. That said, with the Guardrails on the Gateway framework, users can run real-time guardrails on user inputs and LLM outputs to validate and route requests appropriately.

What are the pricing options?

Portkey’s AI Gateway is open-source and can be deployed by just running one command locally - npx @portkey-ai/gateway. This same open source Gateway also powers Portkey’s hosted API that processes billions of tokens for thousands of customers every day, and is free-to-use.

What are the pricing options?

Portkey’s AI Gateway is open-source and can be deployed by just running one command locally - npx @portkey-ai/gateway. This same open source Gateway also powers Portkey’s hosted API that processes billions of tokens for thousands of customers every day, and is free-to-use.

What are the pricing options?

Portkey’s AI Gateway is open-source and can be deployed by just running one command locally - npx @portkey-ai/gateway. This same open source Gateway also powers Portkey’s hosted API that processes billions of tokens for thousands of customers every day, and is free-to-use.

© 2024 Portkey, Inc. All rights reserved

HIPAA

COMPLIANT

GDPR

Transform Your AI app prototypes to be production-ready

Talk to us →