Notes on LLM routing, cost control, reliability, and strict-zero safety.https://ficelle.ai/https://ficelle.ai/blog/provider-breakage-silent-killer/https://ficelle.ai/blog/provider-breakage-silent-killer/Free and free-tier LLM providers rate-limit, retire models, and misreport capabilities, weekly. If your agent points at one provider, it breaks when they do. Routing with cooldowns, quarantine, and fallback is the fix.Sat, 20 Jun 2026 00:00:00 GMThttps://ficelle.ai/blog/strict-zero-routing/https://ficelle.ai/blog/strict-zero-routing/Most "free LLM" setups have a silent paid fallback. Strict-zero removes that risk: free pricing required, tools required, and no paid fallback, full stop. Here is what that means and why it matters.Fri, 19 Jun 2026 00:00:00 GMThttps://ficelle.ai/blog/free-vs-paid-llm-benchmark/https://ficelle.ai/blog/free-vs-paid-llm-benchmark/We ran 76 real agent tasks through free routing and through GPT-5. On classification, JSON extraction, and reasoning, free was at quality parity. Here is where paying still wins, and how much you save when it does not.Thu, 18 Jun 2026 00:00:00 GMT