Auto LLM Ranker – Describe a task in English and get ranked models
Task-specific LLM benchmarking beats generic leaderboards that ignore your actual workload.

Community-ranked free LLM fallback list beats OpenRouter's default, but solves a temporary problem.
Developers building MVPs/demos with free LLM APIs; rapid prototyping workflows.
OpenRouter Free Models Router
- Live model list + reliability rankings (community feedback)
- Filter by use case/health/exclusions via web UI
- await getModelIds() -> dynamic rotation
- GitHub (self-host): https://github.com/kaihochak/free-llm-router
- Hosted (200/day): http://freellmrouter.com
Feedback welcome!
Task-specific LLM benchmarking beats generic leaderboards that ignore your actual workload.
Real-time model health dashboard for free tier switching when your cheap models die mid-session.
Free-tier image agent, but prompt expansion and routing aren't novel vs Comfy UI or ComfyUI agents.
Reference docs for AI agents so they stop hallucinating OpenRouter code, but it's a structured prompt.
Token-efficient DOM-to-list avoids screenshots and vision—works with SPAs, shadow DOM, iframes.
Returns EU Commission consultation numbers for audit proof when VIES alone won't suffice.