My defaultOpenRouterLLM access for client systems

OpenRouter

One SDK, one key, every model. Usually cheaper than running the same open models elsewhere.

In The Stack Weekly: W43: Decision models (Jev, and Cloudflare's Clef), watching · W43: DeepSeek V4.1 Flash 1017, worth a pilot

OpenRouter lets me pick the best cost-per-intelligence model per feature and change my mind later with one string. Same SDK, same key, every provider.

Use it when

Any AI feature in a client system: switch models without changing code.

Pricing

Model prices pass through; OpenRouter adds 5.5% on credit purchases. Example: DeepSeek V4 Pro at $0.66 / $1.98 per million tokens vs $1.32 / $3.96 on Cloudflare Workers AI.

Gotchas, with fixes

Gotcha: Calling it from the browser exposes your API key.

Fix: Always go through a server-side proxy. A Cloudflare Worker is ideal: see the LLM proxy guide.

Provisioning

Pulumi via a Terraform provider · Community Terraform provider

Vendor-published Terraform provider (OpenRouterTeam/openrouter): API keys, guardrails, workspaces, BYOK. Hosted MCP server available.

Instead, consider

Anthropic API (for client systems) Not for client systems.

Worked examples

The LLM streaming proxy

Where it fits

An AI chat or LLM feature: Pick 02.
Every model behind one key. Pick by cost per intelligence.

Sources

OpenRouter

Last reviewed 18 October 2026 by Nicolas Dao

Also in AI

Back to the map →

Want this stack in your AWS account?

Cloudless Labs is a Sydney consultancy. We migrate on-premises systems to AWS, build serverless apps on this stack, and coach teams to ship with Claude Code and Codex.