Cloud LLM router pitching 'unlimited' inference from $10/month — proxies your requests across a pool of upstream models.
Editorial verdict: “Cloud-only LLM router. Useful category, novel pricing. The 'unlimited' math has a known failure mode at heavy usage.”
Which runtime + OS combos this app works against. Source of truth for "will it run on my setup?"
Claudin.io is a cloud routing proxy that sits between your application and a pool of upstream LLM providers (Anthropic, OpenAI, open-weight via OpenRouter-style aggregators). You hit one Claudin.io endpoint with one API key; it picks the model and routes the request. The marketing angle is "unlimited inference starting at $10/month." The category itself is legitimate — OpenRouter, Glama, Together, and Fireworks all play here, and there's real value in a single endpoint that abstracts model selection. Claudin.io's specific positioning is the unlimited-flat-rate pricing, which is unusual in this category (the dominant players are pay-per-token with transparent margins). The pricing model deserves an honest read. Wholesale token costs on Anthropic and OpenAI run $1-15 per million tokens depending on tier. A heavy developer can burn 50-200 million tokens in a month. At those volumes, $10/month is mathematically incoherent unless either (a) heavy users get rate-limited in ways that contradict "unlimited," or (b) the service operates at a loss on heavy users and bets on a 95/5 distribution where light users subsidize. Similar "unlimited AI" proxies have historically either restructured to per-use pricing or shut down within 6-12 months. For someone whose threat model is privacy, cost predictability, or vendor independence, a cloud router is the wrong tool entirely — that's the runlocalai thesis (see /will-it-run for the hardware-side math). For someone who genuinely just wants the cheapest way to access multiple frontier models for casual use, OpenRouter is the more transparent dominant alternative.
Thin SDK / proxy / compatibility layer.
Pre-filled with this app's recommended use case + budget tier. Get the full rig + runtime + model picks.
The full directory — filter by category, runtime, OS, privacy posture, or VRAM.
What this app talks to: Ollama, vLLM, llama.cpp, MLX, LM Studio. The upstream layer.
Did this app work for you on a specific rig? Submit the benchmark — it powers the model + hardware pages.