Launch HN: Tokenless (YC S26) – Automatic model switching to save money
Tokenless, a Y Combinator-backed startup, offers an API router that automatically switches between different AI models to reduce inference costs by up to 50% while maintaining quality. The service works by testing multiple models simultaneously on each request and selecting the most cost-effective one that can handle the task, with users only paying for the selected model. The platform is compatible with OpenAI and Anthropic APIs and claims to deliver equivalent quality to premium models like Claude Opus 4.8 at a fraction of the cost.
Read Full Article →