Kimi K3.1 Pro joins RouterBase for your next code discussionRouterBase now lists Kimi K3.1 Pro and Kimi K3.1 Pro Highspeed. Both bring a 1,048,576-token context window, text and image inputs, and tool support. Highspeed is available for faster output. For a fiOct 26 · 1 min readrouterbase.hashnode.dev
Nano Banana 2.1 integration: separate model pages from API IDsA model page helps a person choose a task. An API model ID tells a service which task to run. When wiring an image feature into an application, keeping these values separate makes configuration and suOct 19 · 2 min readrouterbase.hashnode.dev
Keep 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 “creSep 7 · 2 min readrouterbase.hashnode.dev
Designing 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-Aug 3 · 2 min readrouterbase.hashnode.dev
One 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 it1May 29 · 4 min readrouterbase.hashnode.dev