Self-hosted multi-provider AI gateway routing across 25+ LLM providers through a unified API. 9-stage context pruning, perception pipeline, 9 chat platform integrations. Production systemd service....
Independent AI model benchmarking framework evaluating 339+ models across 14 providers through 12 standardized test phases. Research-grade evaluation infrastructure for AI capability assessment.
25-plugin suite for Claude Code adding session mesh, semantic memory, conductor mode, and tonal continuity. Developer infrastructure for AI-assisted coding workflows.