Gemini, Groq, OpenRouter, Cerebras, OpenAI, HuggingFace — anything OpenAI- or Gemini-compatible. Add a provider in the admin panel, paste the keys, they get tested before they're saved. Dark and…

It logs the last 100 requests with which key succeeded and which one didn't, writes them to logs.jsonl, and gives you an admin panel to add, rename or disable providers without a restart. Keys are…

Every request gets logged: which key succeeded, which ones failed, response time, provider. Last 100 in the admin panel, the rest appended to a JSONL file. Reading logs beats guessing, apparently.

There is also a Telegram bot, so you can poke your models from your phone while pretending to be fully present at dinner. Zero dependencies, self-hosted, plus keep-alive pings so free hosting…

Hit a rate limit on one API key and your whole app falls over. Rotato is the boring fix nobody builds: rotate across multiple keys/providers automatically, with a dashboard so you can actually see…

Comes with an admin panel (dark and light, naturally), logs of your last 100 calls, and a Telegram bot so you can chat with your models from your phone. Zero external dependencies — actually zero.…

There's a clean admin panel (dark + light, because we're not animals) with logs of the last 100 calls, plus a built-in Telegram bot so you can talk to your models from your phone. And zero external…

You also get real-time usage tracking + logs of the last 100 requests, so you can see exactly what's hitting your keys. Bonus: a built-in Telegram bot to chat with your models. And zero external deps…

built a zero-dep node proxy that auto-rotates llm api keys on 429s. openai / gemini / groq / openrouter. shuffles away from recently-failed keys automatically. streaming + admin panel baked in. dms…

built rotato — node proxy that auto-rotates llm api keys on 429s. zero npm deps. point sdk at localhost:8990, it cycles keys silently + logs every call. openai, gemini, groq, openrouter. dms open —…