OpenRouter vs Portkey: How AI Coding Teams Should Choose an LLM Gateway

Forum topic · QianXun · 2026-06-22

Summary

OpenRouter published direct comparison guides (vs Portkey and vs LiteLLM) on June 19, 2026, clarifying how the LLM gateway market is stratifying. OpenRouter is a managed routing network: buy credits, route through one API to 300+ models across 70+ providers with automatic failover, no markup, and a 5.5% platform fee (first 1M requests free). Portkey, acquired by Palo Alto Networks in 2026, is an AI control plane layered over your own provider keys, offering 1,600+ models, governance, prompt management, guardrails, RBAC, and log-based pricing ($49/month production tier). Compliance differs: OpenRouter adds Zero Data Retention and EU routing; Portkey adds HIPAA, BAAs, and VPC hosting. OpenRouter's own threshold: below ~$3,600/month in model spend, use OpenRouter pay-as-you-go; above it, self-host Portkey. The two are also composable, with Portkey treating OpenRouter as an upstream provider.

> Source: OpenRouter Blog, 2026-06-19 / aihot curated, 2026-06-20 > Link: https://openrouter.ai/blog/insights/openrouter-vs-portkey

What Happened

On June 19, OpenRouter published two comparison posts: vs Portkey and vs LiteLLM. Notably, Portkey was acquired by Palo Alto Networks in 2026 and folded into its AI security product line — a key signal of industry consolidation.

Core Positioning

OpenRouter: a managed routing network. Buy credits and route to 70+ providers through one API with automatic failover — no need to manage your own provider keys.
Portkey: an AI control plane. It sits on top of your existing keys and adds governance, prompt management, guardrails, and observability.

Bluntly: OpenRouter helps you call models; Portkey helps you manage the calls.

Model Coverage

Portkey: 1,600+ LLMs via a unified API
OpenRouter: 300+ models across 70+ providers, including 20+ free models for evaluation

The numbers measure different things: Portkey's count is the breadth you *can call* (assuming you hold your own provider keys), while OpenRouter's is the breadth it *actively routes to*, with automatic failover and filtering by price, throughput, latency, and data policy.

Observability and Governance

Portkey centralizes governance: built-in logging, tracing, and dashboards (3-day free / 30-day production / enterprise custom); prompt templates and playground; RBAC; budgets and rate limits; guardrails including PII redaction.
OpenRouter stays light: Broadcast pushes traces to platforms you already use (Datadog, Langfuse, LangSmith, Braintrust, OpenTelemetry, S3); Workspaces provides team organization and budgets.

If you already run a Datadog pipeline, OpenRouter won't reinvent the wheel; if you're building governance from scratch, Portkey works out of the box.

Compliance

Both: SOC 2 Type 2 + GDPR
OpenRouter adds: Zero Data Retention, EU regional routing (trust.openrouter.ai)
Portkey adds: HIPAA, custom BAAs, VPC deployment, data isolation

Medical data leans Portkey; EU data residency leans OpenRouter.

Pricing

| Dimension | OpenRouter | Portkey | |---|---|---| | Model pricing | Zero markup | Bring your own provider key | | Platform fee | 5.5% (5% BYOK; first 1M requests free, $0.80 minimum) | Billed per log | | Free tier | First 1M requests/month | Developer tier, 10K logs/month (not for production) | | Production | Usage-based | $49/month + 100K logs (+$9 per extra 100K) | | Enterprise | — | Custom (HIPAA, SSO, private deployment) | | Self-hosting | — | Free open-source gateway (12K+ GitHub stars) |

Decision Threshold (OpenRouter's Own Advice)

> Monthly model spend ≥ $3,600 (≈ $200/month infrastructure) → self-host Portkey > Monthly model spend < $3,600 → OpenRouter pay-as-you-go

Not mutually exclusive: Portkey can treat OpenRouter as an upstream provider (governance + routing combined); OpenRouter plans to list Portkey as a Broadcast target (traces flowing the other way).

OpenRouter vs LiteLLM (Same-Day Companion Post)

OpenRouter: hosted on Cloudflare's edge, 5.5% platform fee (first 1M requests free), automatic failover.
LiteLLM: self-hosted proxy (Docker/PostgreSQL/Redis); data never leaves your network; free and open source, but you bear infrastructure costs (a production deployment runs a few hundred dollars/month).

LiteLLM suits compliance scenarios where data must never leave the internal network.

Why It Matters

1. AI coding infrastructure is maturing: LLM gateways are clearly stratifying — routing network (OpenRouter) / control plane (Portkey) / self-hosted proxy (LiteLLM). Each layer has its own customer. 2. Consolidation signal: Portkey's acquisition by Palo Alto shows enterprises are taking the AI gateway market seriously; similar acquisitions may follow. 3. Pricing transparency: 5.5% platform fee vs per-log billing vs free self-hosting — how AI coding teams do the math is diverging. 4. Compliance is fragmenting: EU data residency, HIPAA, and Zero Data Retention are now key selection criteria, not afterthoughts. 5. A composable ecosystem: OpenRouter + Portkey isn't either/or — teams can start with a single gateway and layer on governance as they scale.

Risks and Open Questions

Portkey's independence post-acquisition: will the open-source version stay active under Palo Alto? Will pricing change?
OpenRouter's 5.5% cumulative cost: long-term, a self-hosted proxy like LiteLLM may be cheaper — if your team can operate it.
Steep governance costs of multi-provider setups: moving from one provider (OpenAI/Anthropic) to many causes a step-change, not a linear increase, in governance and observability costs.
Quality consistency under automatic failover: how do you handle behavioral differences when routing across providers? OpenRouter's Auto Router uses NotDiamond for model selection, but real-world behavioral variance remains an open problem.