Router that classifies coding/documentation tasks with a local model and dispatches to the cheapest/best-fit open-weight model across OpenRouter and Neuralwatt Cloud, weighted by cost, ecological impact, and per-category proficiency. Includes: - schema.sql (models, proficiency, energy_observations tables) - poller.py (OpenRouter + Neuralwatt catalog sync) - config.py/config.yaml (pydantic-validated weights/thresholds) - scoring.py (cost/eco/proficiency/composite normalization) - tiering.py + tier.py (heuristic tier resolver + apply_tiering DB pass) - tests/ (28 tests, all green)
67 B
67 B