Automatic LLM router — 82% cost savings, 79.4% accuracy, 93.4% pass rate. Drop-in OpenAI proxy.

Open-source LLM router & AI cost optimizer. Routes simple prompts to cheap/local models, complex ones to premium — automatically. Drop-in OpenAI-compatible proxy for Claude Code, Codex, Cursor, OpenClaw. Saves 40-70% on AI API costs. Self-hosted, no middleman.

Provider-pluggable orchestration runtime for multi-model AI inference. ( Sakana Fugu style )

Route complex Codex tasks to model-specific background workers with bounded concurrency and lead-agent verification.

Private AI assistant, AI agent, and unified model proxy for Claude Code, Codex CLI, Gemini CLI & OpenClaw. Skills, MCP, tools, channels, tasks,model routing, accounts, keys, logs, dashboard.

Task-level model routing through persistent subagents, with local-first GUI and CLI orchestration

Route inference across providers.

在自己的 Mac 上搭一套常驻、自愈、走订阅、墙内也能用的自托管 AI 伴侣 · 人看版讲思路，机看版给完整规格 · 文 / 小C & Grace

Free, self-hosted AI model router. OpenRouter / ClawRouter alternative using your own API keys. 14-dimension classifier routes to the right model (Anthropic/OpenAI/Kimi) automatically. No middleman, no markup. Built for OpenClaw.

TRINITY in Elixir (An Evolved LLM Coordinator): route LLM calls via a small-model hidden-state router + Axon coordination head, with Thinker/Worker/Verifier orchestration and policy loop for acceptance-driven completion.

OpenVitamin is a local-first AI execution platform that unifies Agents, Workflows, and multi-model inference into a single programmable system — designed for building real, production-grade AI applications.

The Python Harness for Production AI Multi-Agent Systems

Turn your strongest Claude model into a team-lead orchestrator: it plans and reviews while cheaper Claude or Codex workers execute — with blind verification before anything counts as done.

Auto-delegation, status truth, and cost-aware model routing for OpenClaw agents

Claude Code hooks that auto-switch model tier based on task complexity

🏆Winning Project | ModelGate is a contract-aware AI control plane that ingests customer contracts, extracts SLA/privacy/routing constraints, and generates an OpenAI-compatible endpoint that automatically routes every request to the optimal model. Simple queries go to cheap models. Complex queries go to premium ones.

Main agent plans and reviews; cheaper sidekick edits. Devin Fusion pattern for OpenCode

把多模型 Provider、本地密钥和 HeavySkill 讨论统一成 MCP 网关 | Local MCP gateway for multi-provider LLM routing, secrets, and HeavySkill discussions.

Open-source Fugu: the open-source LLM orchestration brain.

Stateless LLM runtime that dynamically routes, loads, executes, and unloads models per request with bounded VRAM caching and intelligent model selection.