OpenRouter

OpenRouter is a unified API gateway for large language models. It places one OpenAI compatible endpoint in front of hundreds of models from many providers, so developers use a single API key, a single bill and a single SDK rather than integrating each vendor separately. It is built for teams shipping AI features who want to switch models or compare them on the same prompt without code changes. Automatic fallbacks route around provider outages and rate limits, and cost tracking spans providers. Beyond text and chat, the platform covers image, video, speech to text and text to speech models, each with its own endpoint. Supporting resources include a model comparison page covering pricing, context length and benchmarks, a machine readable model API, and rankings by usage. The vendor also publishes an MCP server and documentation aimed at coding agents. Pricing is pay as you go per token, with no markup on the provider's list price on any plan. Platform fees apply when credits are bought: none on the Free plan, 5.5 percent on Standard and 8 percent on Business, with negotiated fees on Enterprise. The Free plan offers 25 or more free models, limited to 20 requests per minute and 50 per day, rising to 1,000 per day once at least $10 of credits has been purchased. Standard lists 500 or more models across 80 or more providers. Business adds in region routing for the EU and US and up to 1,000 workspaces. Bring your own key usage is free up to $25,000 of list price inference per month on Standard and Business, then 5 percent. Enterprise is priced by contract and adds SSO, SAML, contractual SLAs, invoicing and a dedicated support team.