All › AI › Portkey

Portkey

AI gateway with observability and guardrails

Portkey provides an AI gateway for routing, caching and fallback across LLM providers, plus logging, prompts and guardrails.

Visit Portkey ↗

Pricing: Free plan

Platforms: Web, API, Self-hosted

Categories: AI, Developer Tools

Portkey compared with similar tools

Product	Price	Open source	Self-host	Platforms
Portkey	Free plan	Yes	Yes	Web, API, Self-hosted
Langfuse	Free plan	Yes	Yes	Web, Self-hosted, API
Helicone	Free plan	Yes	Yes	Web, Self-hosted, API
LiteLLM	Free plan	Yes	Yes	Self-hosted, API
Braintrust	Free plan	–	–	Web, API
Promptfoo	Free plan	Yes	–	Mac, Linux, Windows, Web
Dify	Free plan	Yes	Yes	Web, Self-hosted

Similar to Portkey

Langfuse

Open-source LLM engineering platform for tracing and evals

Langfuse provides tracing, prompt management, evaluations and datasets for LLM applications, cloud or self-hosted.

Free plan · Web, Self-hosted, API · Website ↗

Helicone

Open-source LLM observability and gateway

Helicone logs and analyzes LLM requests for cost, latency and quality, and offers an AI gateway, self-hostable or managed.

Free plan · Web, Self-hosted, API · Website ↗

LiteLLM

Open-source LLM gateway and SDK for 100+ providers

LiteLLM is a Python SDK and proxy server that calls many LLM providers through an OpenAI-style API, with budgets, keys and logging.

Free plan · Self-hosted, API · Website ↗

Braintrust

Evals, logging and prompt playground for AI products

Braintrust helps teams evaluate AI models and prompts, log production traces, and iterate in a shared playground.

Free plan · Web, API · Website ↗

Promptfoo

Test and red-team LLM apps from the command line

Promptfoo is an open-source CLI for evaluating prompts and models with test cases and for red-teaming LLM apps for security issues.

Free plan · Mac, Linux, Windows, Web · Website ↗

Dify

Open-source platform for building LLM apps and agents

Dify provides a visual workflow builder, RAG pipelines, agent tools and model management for building AI applications.

Free plan · Web, Self-hosted · Website ↗

llama.cpp

LLM inference in C/C++ on everyday hardware

llama.cpp is an open-source C/C++ engine for running LLMs in GGUF format on CPUs and GPUs, with a built-in server.

Free · Mac, Windows, Linux · Website ↗

vLLM

High-throughput LLM inference and serving engine

vLLM is an open-source library for fast LLM serving with PagedAttention, continuous batching and an OpenAI-compatible server.

Free · Linux, Self-hosted · Website ↗

Questions

Is Portkey free?

Portkey is free to start, with paid plans.

Is Portkey open source?

Yes, Portkey is open source.

Can Portkey be self-hosted?

Yes, you can run Portkey on your own server.

What platforms does Portkey run on?

Portkey is available for Web, API, Self-hosted.

What are some alternatives to Portkey?

Similar indie products include Langfuse, Helicone, LiteLLM, Braintrust and 4 more.

Is this your product? Submit an update.