Launching today
OneMux
Low-cost, OpenAI-compatible API for multiple AI models
13 followers
Low-cost, OpenAI-compatible API for multiple AI models
13 followers
OneMux is a low-cost AI API gateway for indie hackers and small teams. Access multiple AI models—including GPT, Claude, Gemini, and more—through one OpenAI-compatible API. Switch models without rewriting your integration and manage usage in one place. Launch offer: pay $10 and get $100 in API credits—up to 90% off launch credits.
Would love a built-in cost calculator that lets me estimate monthly spend by model before committing, since switching between GPT and Claude can get pricey fast.
@mirayj8ur Great suggestion! OneMux already tracks usage by model, including input/output tokens, cache usage, latency, and per-request cost, so you can see exactly what each model is costing.
We don’t have a forward-looking monthly spend estimator yet, but model-level forecasting based on expected usage would be a useful next step. Thanks for the idea!
Would love to see built-in cost analytics broken down by model and prompt, maybe with the ability to set per-team or per-key spending caps. Right now managing usage across multiple clients feels like guesswork, and having a dashboard that flags when you're approaching a threshold would save a lot of headaches for small teams like ours.
@ariflq7p Thanks, Arif — that’s a very useful use case.
OneMux already tracks model-level token usage and per-request cost. We’re also evaluating more granular controls for per-key or per-team spending limits.
Are you currently managing multiple AI providers in production, or are you still evaluating options? If you’d like to test OneMux, new users can start with $10 in free API credits: https://onemux.net
Switched between Claude and GPT mid-project without touching my code, and the usage dashboard finally makes sense of my spending. The launch credits are a solid way to test it out too.
@hatunasralh5mi Glad to hear that! Which models did you switch between, and what were you building? We’d love to learn more about your workflow.
the launch credit deal is super tempting, especially for someone like me juggling a few side projects. one thing that would really help: built-in cost tracking per project or feature flag, so i can see exactly which model is eating my budget. right now i have no idea if claude is costing me more than gpt for my use case
Switched a side project from GPT to Claude through OneMux in about two minutes, no code changes needed. The unified usage dashboard is genuinely useful for keeping track of spend across models.