ML Junction: One API for every AI model, with a receipt per request

Hi everyone, we are the team behind ML Junction. Every AI model has its own rules: which settings it takes, how you turn thinking up or down, how tool calls must look. Change the model and your code breaks. We built ML Junction so you write the request once. It translates the request for the model it lands on, routes it by capability, provider health, privacy and cost, falls back to another provider if one fails, and returns a receipt that shows what ran and what it cost. It works with the official OpenAI and Anthropic SDKs by changing the base URL. You pay provider rates at cost plus one routing fee, on prepaid credits or your own provider keys. We would like to hear which model quirks bite you most.