--- title: Benchgen emoji: 📊 colorFrom: yellow colorTo: red sdk: static pinned: false ---
**Nobody can prove an AI agent actually works.** Benchgen helps you discover benchmarks, run evaluations on any model or agent, find where and why it fails, and check the fix with the same tests before and after. ## What we publish here | Release | What it is | Link | |---|---|---| | **Router Lite** (model) | A 10,245-parameter linear router head on a frozen Qwen3-1.7B. Picks which of 5 models should answer a query | [benchgen/benchgen-router-lite](https://huggingface.co/benchgen/benchgen-router-lite) | | **Router Pilot rewards** (dataset) | Per-question accuracy of 5 agents over 3 attempts (46 tasks) | [benchgen/router-pilot](https://huggingface.co/datasets/benchgen/router-pilot) | | **Router Pilot tasks** (dataset) | 1,110 normalized questions from public benchmarks, with per-source licences | [benchgen/router-pilot-tasks](https://huggingface.co/datasets/benchgen/router-pilot-tasks) | Every card states its method, licence, limits and a link to the relevant benchmark on Benchgen. ## Run it yourself 1. Open the [Router Fidelity Benchmark](https://benchgen.com/benchmarks/platform/router-fidelity-benchmark?utm_source=huggingface&utm_medium=org_card&utm_campaign=org&utm_content=router-fidelity) (sign-in required to run). 2. Register your own model or router head, pick it on the Evaluate tab and run. 3. Compare with our published run. The [Router Lite guide](https://benchgen.com/docs/guides/router-head/benchgen-router-lite?utm_source=huggingface&utm_medium=org_card&utm_campaign=org&utm_content=guide) has the steps. ## Honest limits Our releases are pilots. Router Lite was trained on 46 labelled tasks, and its results are directional, not a large-sample benchmark. Each card says exactly what the numbers do and do not show.