One API key and one endpoint in front of hundreds of models — switch model without switching integration.
Pick this whenYou want to change models freely (or fall back automatically) without rewriting credentials in every workflow.
What makes it different
Rankings and side-by-side model comparison pages that show what's actually being used
A single credential in an n8n agent node covers every model it fronts
Makes model choice a runtime decision rather than an architectural one
What it does
Routes requests to hundreds of hosted models through one API
Exposes rankings and price/perf comparisons for picking
Serves as the chat-model credential inside automation platforms
What it does not do
Common wrong expectationsDoesn't host models itself — it's a router, so upstream outages are still outages
Doesn't beat a direct vendor key on price for a single model you've settled on
Where it sits in a stack
The model-access layer of any agent build where you expect to experiment, benchmark, or switch as prices move.
ShelfModels & model access
Our status · watchNothing in the pipeline calls models by API yet.
For your projectsThe right first stop if the SQLite/vector phase needs an embedding or summarization API — one key, easy to benchmark alternatives against.
| Session | Coverage | Moment | What happened |
|---|
| Basecamp 3: Introduction to n8n — I | demonstrated | 1:35:00 | Rankings and model-compare pages toured; positioned as the single-API model switcher for agent nodes |
| Session 29: Build Always-On Assistants — Autonomy Is Designed, Not Switched On (OpenClaw on a VPS, Hermes on the Desktop) | demonstrated | | API key as OPENROUTER_API_KEY environment variable; usage page shown with per-model spend. |
| Day 1: APIs, Webhooks, How LLMs Predict, CoStar vs Meta Prompting, Local Models, and a Live RAG Build | demonstrated | | 606 models; run several in parallel to compare |
| Day 3: n8n From Zero — Triggers, Fixed vs Expression, an AI Spam Classifier, and Your First Lead-Routing Workflow | demonstrated | | Rankings tab by category (office hours) |
| Day 2 (morning): Agentic Workflows in Code - CrewAI from One Agent to a DevOps Crew to a Parallel Investment Adviser with Custom Tools | demonstrated | | One key for 300+ models |
| Day 1: The Magic Prompt Formula, a Toolkit Tour, Thinker-and-Doer Data Analysis, the Blind-Men-and-the-Elephant Research Method, and a Living Pitch Deck | demonstrated | | 517 models, rankings by category |
| Day 2: Custom GPTs and Gems, Markdown Prompting, Cloning Your Writing Style, MCP in Daily Use, a Vibe-Coded Calorie Counter, and the Consultant-to-SaaS Road | demonstrated | | 500+ models |
| AI Sprint: Claude Code with Antigravity — Day 1 (Taste as the Moat, the Muscle-and-Mirror Setup, and the Model Council Build) | demonstrated | | The council's model layer; live catalog checks correct the agent (Gemini 3.1 Pro, Grok 4.2); ':online' suffix adds web search; $0.26 total |
| AI Sprint: Claude Code with Antigravity — Day 2 (CLAUDE.md, the 50% Context Rule, and the Sketch-to-App Build) | demonstrated | | User-selectable vision models behind the key-protecting server route |
| AI Sprint: The Hermes Agent — Day 1 (Always-On Agents, the Employee-on-Day-One System, and the Self-Improving Ghostwriter) | demonstrated | | Hermes' provider (Opus 4.6 for price); also the image-generation route; credits checked mid-debug |
| Understanding Jev, the AI Built for Decisions (Sprint Day 2): Two Second-Brain Systems - a Leadership Self-Review Built Problem-First, and a Tag-Classifying Knowledge Base Where a Cheap Decision Model Replaces a Chat Model | demonstrated | | Model prices and calls to Jev and Haiku |
| AI Sprint: The Open Source Playbook — Day 1 (Licenses, Open Weights, and Why Open Isn't Free) | demonstrated | | BYOK layer for Open Design; the activity page as the cost meter |
| AI Sprint: The Open Source Playbook — Day 2 (Weights 101, OpenCode, and the Subsidy Math) | demonstrated | | The BYOK spine: provider config in OpenCode, per-model activity as the meter |
| AI Sprint: AI Evals & Reliability — Day 2 (Testing Systems That Never Answer Twice) | demonstrated | | BYOK keys powering the support bot's multi-model orchestration |
| AI Sprint: Windsurf & Devin — Day 2 (Building Superplexity: Plan with Opus, Execute with Cascade, Review with Devin) | demonstrated | | Sonnet 4.6 for synthesis, Sonar for search, keys in .env — 'I won't face rate limits because I'm using OpenRouter' |
| Office Hours III: Backing Up Claude Chats with Obsidian, Bridging Chat and Cowork, What Fable Is For, and the Context-Rot Problem | explained | | Hosted open models for the VPS agent |
| Day 10: Principles for Building Agents — Three Levels of AI, the 7-Question Workflow-vs-Agent Test, Goldilocks Scope, Agent Prompting, and Tool Design | explained | | Unified key and pricing comparison |
| The Universal Research Framework: SMART-AI Problems, the Research Tree, Three Source Layers, RECAP Evaluation, the Insight Pyramid, and Packaging | explained | | Model-usage rankings per use case |
| AI Sprint: The Hermes Agent — Day 2 (Loop Engineering: the Scheduling Loop, the Delegation Loop, and the Skill Auditor) | explained | | The catalog for the cheap-model lane (DeepSeek, Kimi, open models); fallback chains |
| Community Session II: The Meta-Harness — Praful Thakkar on AgentGrid, Orchestrator-plus-Workers on One Canvas, Hand-Off Between Models, and OAuth Instead of API Keys | mentioned | | Optional key source |
| Office Hours IV: PostHog over Plausible, the Launch-Fast Loop, Scraping Ethics, Agent Swarms with Memory, and Why You Never Give Credits Away Free | mentioned | | Compare benchmarks, pricing, latency |
| Office Hours XI: rolecompass.ai's Launch, Redesigning a Client Site Without Losing Its SEO, the Hermes Harness Tour, and Turning a Workflow into a Skill | mentioned | | Reported Stripe acquisition |
| Mentor Session XV: Selling Beats Building - Paul's Referral-Network Platform, Phone-Book Outbound, Give-First Communities, One Use Case Not a Hundred, and 'Know Thyself' | mentioned | 0:44:00 | 'OpenRouter for MCPs' positioning |
| Day 1: How a Model Learns (Teachable Machine Live), Context Engineering, CO-STAR, Artifacts, Connectors, and Claude Cowork Running a Daily Stand-up | mentioned | | Bring-your-own-key model access |
| Day 1: What Makes an LLM an Agent - Planning, Tools, Memory, Actions; Five Design Patterns; and Cursor, Antigravity and Claude Code Live | mentioned | | ~600 models listed |
| Weekly AI Updates - October 2026 digest (Oct 7 so far: Gemini 4 'Argon', America.gov AI, Meta Muse, NVIDIA OpenShell, Grok Bot rewards, Copilot teammates, Claude Sonnet 5.5 deep dive; Mark Palmer's GBox showcase; Oct 14 so far: Google Health Coach, the White House frontier-AI accord, OpenAI Dev Day 2026, Anthropic's Oct 8 usage-policy update, Hark, Microsoft MAI voice models, Claude Haiku 5.5) | mentioned | | |
OpenRouter