Show HN: ModelPlane, one OpenAI-compatible gateway for AI models

ModelPlane, a new OpenAI-compatible gateway, lets developers route AI requests across OpenAI, Anthropic, DeepSeek, OpenRouter, and custom services through a single API, with features like model groups, fallback routing, and usage billing. The service offers 1 million free tokens without a credit card, and paid plans start at $0.02 per 1 million tokens.

Every model. One endpoint. Route across OpenAI, Anthropic, DeepSeek and more behind a single API. Model groups, fallback routing, and usage billing - built in. No credit card · First 1M tokens free Six primitives. One gateway. Each works on its own. Together they replace the model-routing layer for every app you ship. Model Groups Package backends, models and routing policy behind one stable name your app calls directly. Routing Single-backend, weighted load-balancing, fallback, or conditional routing by request metadata. Multiple Providers OpenAI, Anthropic, DeepSeek, OpenRouter and any custom OpenAI-compatible service. Coding Plans Run coding-plan backends beside credit-based keys, with quota and balance visibility. Thinking Output One reasoning request format; read standardized reasoning text across every model. Usage & Billing Track spend and tokens per group, key and customer tier in a single console. Failover before your users notice. Define a policy once. The gateway evaluates weights, retries, and fallbacks on every request — and hands back a standardized response no matter which backend answered. Pay for credits, not seats. All plans draw credit at the same $0.02 / 1M-token rate. Compare all plans /pricing/ $1 in credit to start · 1 active key · $0.02 / 1M tok $20 credit included monthly · 10 active keys · $0.02 / 1M tok Start shipping. One endpoint for every model. Free for the first 1M tokens — no credit card required.