ai-gateway
Portkey
Enterprise AI gateway + observability + guardrails + prompt mgmt
AI gateway with observability, guardrails, caching, fallbacks, prompt management. Universal API for 250+ LLMs. OSS gateway + managed cloud.
Pricing
| Tier | Price | Notes |
|---|---|---|
| Developer (Free) | Free | Free forever. 10k logs/month. Universal API + key management. 3 prompt templates. Basic observability. |
| Gateway (OSS) | Free | MIT-licensed gateway only (no observability UI). Self-host for routing/fallbacks. |
| Production | $49/mo | $49/month. 100k logs ($9 per additional 100k). Fallbacks, load balancing, retries, semantic caching. Unlimited prompts. RBAC. |
| Enterprise | Custom | Custom. 10M+ logs/month. Custom guardrails, advanced evals, SSO, budget controls, VPC + on-prem, SOC2, HIPAA, GDPR. |
Limits
| Tier | Metric | Value | Notes |
|---|---|---|---|
| — | compliance | SOC2 Type 2, GDPR, HIPAA on Enterprise | Compliance certs |
| — | gateway latency | <100ms added latency | Gateway overhead |
| — | guardrails | Custom guardrail hooks on Enterprise | Guardrails |
| — | open source gateway | Gateway is MIT-licensed — can self-host without cloud dependency | OSS component |
| — | prompt versioning | Version + test + deploy prompts centrally | Prompt mgmt |
| — | providers supported | 250+ LLMs: OpenAI, Anthropic, Google, AWS Bedrock, Azure OpenAI, Groq, Together, Mistral, Cohere, etc. | Providers |
| — | semantic caching | Caches on meaning not just exact match — Production+ | Semantic caching |
| — | virtual keys | Per-user/app virtual keys with budget caps + rate limits | Virtual keys |
Features
- AI Gateway — Unified OpenAI-compatible API to 250+ LLMs.
- Alerts — Thresholds on latency, error rate, cost, usage.
- Budget Controls — Per-key + per-team spending limits.
- Evaluations — Built-in evaluator templates + custom.
- Fallbacks — Config-driven provider fallback chains.
- Guardrails — Pre/post processors for safety + compliance.
- Load Balancing — Round-robin, weighted, least-latency across providers.
- MCP Support — Use MCP servers as tools through gateway.
- Observability — Logs, traces, feedback, alerts, cost tracking.
- OSS Gateway — Open-source gateway (portkey-ai/gateway).
- Prompt Library — Shared prompt library + public marketplace.
- Prompt Templates — Version + test + collaborate on prompts.
- Retries — Configurable retry policies per route.
- Role-Based Access Control — Team permissions on prompts + keys.
- Semantic Caching — Vector-based cache on query meaning.
- Simple Caching — Exact-match cache.
- Virtual Keys — Per-app keys with budget + rate limits + permissions.
- VPC Deployment (Ent) — Deploy in your own VPC for compliance.
Developer interfaces
| Slug | Name | Kind | Version |
|---|---|---|---|
| sdk-node | portkey-ai (Node) | sdk | 1.x |
| sdk-python | portkey-ai (Python) | sdk | 1.x |
| rest-api | Portkey API (OpenAI-compat) | rest | v1 |
| cli | Portkey CLI | cli | 0.x |
| dashboard | Portkey Dashboard | other | — |
| mcp | Portkey MCP | mcp | — |
Compare Portkey with
ai-api
Portkey vs Anthropic API
Side-by-side breakdown.
ai-api
Portkey vs AssemblyAI
Side-by-side breakdown.
ai-api
Portkey vs Deepgram
Side-by-side breakdown.
ai-api
Portkey vs ElevenLabs
Side-by-side breakdown.
ai-api
Portkey vs Google Gemini API
Side-by-side breakdown.
ai-api
Portkey vs Groq
Side-by-side breakdown.
ai-api
Portkey vs OpenAI API
Side-by-side breakdown.
ai-api
Portkey vs Replicate
Side-by-side breakdown.
Staxly is an independent catalog of developer platforms. Outbound links to Portkey are plain references to their official pages. Pricing is verified at publication time — reconfirm on the vendor site before buying.