Honest pros, cons, and verdict on this llm gateways & infrastructure tool
✅ Provider-agnostic routing with declarative configs that an SRE can change without app deploys
Starting Price
Free
Free Tier
Yes
Category
LLM Gateways & Infrastructure
Skill Level
Developer
AI gateway and control plane for production GenAI: routes calls across 250+ LLMs with one unified API, plus guardrails, prompt management, observability, budgets, and an MCP-aware agent runtime.
Portkey is an opinionated AI control plane that sits between your application and your LLM providers. The core is an AI gateway: a single endpoint that proxies to 250+ models from OpenAI, Anthropic, Google, Mistral, AWS Bedrock, Azure OpenAI, Cohere, Together, Groq, Fireworks, and others, with config-driven routing rules, fallbacks, load balancing, semantic caching, and retries. Around that gateway, Portkey layers a prompt library with versioned templates, an evals system for offline and online testing, guardrails (PII, jailbreak, toxicity, JSON schema, custom checks), and full request-level observability with cost and latency dashboards. For enterprises, Portkey adds RBAC, virtual keys, per-team budgets and rate limits, and SOC 2 / HIPAA / GDPR posture, plus self-hosted and BYOC deployments. The newer Agents product treats the gateway as the runtime for MCP-style tool-using agents: you register tools (including MCP servers), models, and policies once, then any agent in the org can call them through Portkey with central logging and budget enforcement.
per month
per month
Unified API marketplace giving developers a single OpenAI-compatible endpoint and one bill for 300+ models from every major and minor LLM provider.
Starting at Free
Learn more →Cloudflare AI Gateway accelerates AI applications with intelligent caching, automates cost optimization through rate limiting, and analyzes LLM usage across OpenAI, Anthropic, Google providers. Reduce AI costs 60%+ with response caching. Free tier available.
Starting at Free
Learn more →Open-source LLM observability and AI gateway — logs every prompt, response, cost, and latency across 20+ providers with a one-line proxy or async SDK, plus caching, retries, and prompt experiments.
Starting at Free
Learn more →Portkey delivers on its promises as a llm gateways & infrastructure tool. While it has some limitations, the benefits outweigh the drawbacks for most users in its target market.
AI gateway and control plane for production GenAI: routes calls across 250+ LLMs with one unified API, plus guardrails, prompt management, observability, budgets, and an MCP-aware agent runtime.
Yes, Portkey is good for llm gateways & infrastructure work. Users particularly appreciate provider-agnostic routing with declarative configs that an sre can change without app deploys. However, keep in mind pricing page is js-rendered and dollar amounts must be confirmed manually on site.
Yes, Portkey offers a free tier. However, premium features unlock additional functionality for professional users.
Portkey is best for Multi-provider production stacks that need one consistent API and automatic fallback routing and Platform teams centralizing budgets, RBAC, and logging across many internal AI apps. It's particularly useful for llm gateways & infrastructure professionals who need advanced features.
Popular Portkey alternatives include OpenRouter, Cloudflare AI Gateway, Helicone. Each has different strengths, so compare features and pricing to find the best fit.
Last verified March 2026