RRouterBaseinrouterbase.hashnode.dev·Sep 7 · 2 min readKeep H3 Max Model Selection Separate From Your Creative BriefRouterBase now lists MiniMax H3 Max for text-to-video and image-to-video. A useful application design keeps the user’s creative brief separate from the API model identifier. The user might choose “cre00
RRouterBaseinrouterbase.hashnode.dev·Aug 3 · 2 min readDesigning an Async Video Pipeline for MiniMax H3’s Three Input ModesVideo generation should be modeled as a job system, not a long HTTP request. That becomes especially important when one model supports several ways to start. RouterBase now exposes MiniMax H3 as text-00
RRouterBaseinrouterbase.hashnode.dev·May 29 · 4 min readOne Base URL, 200+ Models: Why I Stopped Installing a New SDK for Every AI ProviderIf you've shipped anything with LLMs in the last year, you've probably lived this: you start on OpenAI, then a teammate wants to try Claude for long-context work, then someone benchmarks Gemini and it10