OpenRouter
OpenRouter is a unified API over many LLM providers. Rechecked 2026-08-17: openrouter.ai/docs/faq says it passes through provider list prices with no inference markup, then charges a 5.5% fee ($0.80 minimum) on Stripe credit purchases and 5% on crypto. openrouter.ai/pricing still lists a Free row (25+ free models, 4 free providers) and Enterprise. BYOK has a fee-free allowance of $25,000 / month list-price inference on pay-as-you-go and $200,000 / month on Enterprise, then 5% of the equivalent OpenRouter price. Free-model rate limits in the FAQ source are 50 requests/day with no credits and 1,000 / day once you have at least $10 in credits. Model IDs change; do not freeze GPT-4o / DALL-E as the catalog.
Compare Ollama if you wanted a local runner plus Ollama's own cloud, and LangChain if you only needed a client library.
Key Features
- One key, many providers: The product is routing and failover, not a single vendor model.
- Credits wallet: You deposit, then burn per-model token (and sometimes per-request / image / reasoning) prices shown on each model card.
- BYOK: Use your own provider keys after the monthly free allowance. The 5% overage is deducted from OpenRouter credits.
- Free tier: Limited free models. It is not unlimited GPT-4o.
Use Cases
- Teams that want one invoice across Anthropic, OpenAI, Google, and others.
- People who will BYOK after they pass $25k / month list price.
- People comparing OpenRouter vs Ollama on router vs local runner.
Limitation: We did not re-verify every model dollar on the marketplace. Quote the live model card. Domain rating 142 is leftover catalog metadata, not a new audit.
Pricing
openrouter.ai/pricing and FAQ on 2026-08-17.
| Path | Price | Notes |
|---|---|---|
| Free | $0 | 25+ free models, 4 free providers. 50 RPD without credits; 1,000 RPD with $10+ credits. |
| Pay-as-you-go credits | Provider list price | 5.5% Stripe purchase fee ($0.80 min). 5% crypto. |
| BYOK | $0 fee to $25k / month list | Then 5% of the equivalent OpenRouter price. Enterprise allowance $200k / month. |
| Enterprise | Sales | Higher BYOK allowance and extras on the pricing page. |
If a leftover blog still lists only GPT-4o / Claude / Gemini / DALL-E with no fee, treat this table as the source.
Getting Started
- Create an account at openrouter.ai and read one live model card before you buy credits.
- Add credits on the Credits page if you need paid models.
- Recheck BYOK only after you understand the $25k / 5% line.
- Compare Ollama if you wanted weights on your own GPU.
Frequently Asked Questions
Does OpenRouter mark up tokens?
FAQ says no markup on provider inference. The fee is on credit purchase and on BYOK overage.
Is there a $20 Pro seat?
Not on the page we opened. Billing is credits plus optional Enterprise.
Same as LangChain?
No. LangChain is a client framework. OpenRouter is an HTTP router and wallet.
Alternatives
Tips
- Quote 5.5% Stripe and $25k BYOK, not "no fees."
- Do not paste GPT-4o as the current catalog.
- Recheck the live model card before you put a dollar in a contract.
Conclusion
OpenRouter is a credit-metered multi-provider router with a documented purchase fee, not a GPT-4o landing page. Start at openrouter.ai/pricing, then decide whether Ollama already covers the models you need.
Comments
No comments yet. Be the first to comment!
Related Tools
9Router
9router.com
Free MIT AI router and token saver that connects Claude Code, Codex, Cursor, and Cline to 40+ providers with auto-fallback and RTK token compression.
Cloudflare AI Gateway
www.cloudflare.com/developer-platform/products/ai-gateway
LLM proxy on Cloudflare. Rechecked: core features free. Paid logs 10M/gateway; extra requests $0.05/million. Not an unmetered endpoint.
CheaperInference
cheaperinference.com
CheaperInference resells discounted AI inference through one OpenAI-compatible API, cutting model costs by up to 30% with usage-based billing, no contract.
Related Insights
After I Connected Obsidian to OpenClaw, It Started Helping Me Make Decisions
Once Obsidian stopped being just a place to store notes and started working with OpenClaw, it began helping me organize context, connect information, and improve real decisions.

Anthropic Subagent: The Multi-Agent Architecture Revolution
Deep dive into Anthropic multi-agent architecture design. Learn how Subagents break through context window limitations, achieve 90% performance improvements, and real-world applications in Claude Code.
Stop Cramming AI Assistants into Chat Boxes: Clawdbot Picked the Wrong Battlefield
Clawdbot is convenient, but putting it inside Slack or Discord was the wrong design choice from day one. Chat tools are not for operating tasks, and AI isn't for chatting.