model-switching

One dashboard for all your local AI coding agents. Switch providers, manage sessions, and orchestrate tasks across Codex, Claude Code, Gemini CLI, CodeBuddy Code, Pi, OpenCode, KiloCode, and OpenClaw. Zero cloud, local-first control plane.

Production LLM call layer for AI agents and tools: keep OpenAI/Anthropic/AI SDK/LiteLLM, hot-swap models with MDA presets, and add cache, retries, circuit breakers, key rotation, singleflight, and Python/TypeScript/Rust parity.

A plugin for ASReview allowing for switching between models during runtime.

A focused desktop provider manager and router for AI coding tools.

⚙️ Advanced fallback plugin for OpenCode agents — model switching, retry with backoff, and context overflow handling

MCP server for hot-swapping llama.cpp models in Claude Code - launchctl (macOS) + systemd (Linux)

[TCCN 24] GAIKube: Generative AI-based Proactive Kubernetes Container Orchestration Framework for Heterogeneous Edge Computing

Cross-platform Codex model switcher for OpenAI-compatible providers and local llama.cpp models.

Lightweight web UI for llama.cpp with dynamic model switching, chat history & markdown support. No GPU required. Perfect for local AI development.

RAMS: Resource-Adaptive Model Switching for Edge AI - dynamically switches YOLOv8 tiers based on system resource pressure with a safety override for vulnerable road users

Anthropic-compatible Claude Code proxy for OpenCode Zen and OpenAI-compatible chat completions models.

Easily switch the AI provider and model that Claude Code uses (OpenRouter, Z.ai/GLM, Ollama, HuggingFace, Free-Way) across all your projects with one command. Free-tier focused, secrets-fre

Hermes Agent plugin: auto model switching based on token quota and content type detection. 根据额度和内容类型自动切换模型，对话不中断。

A lightweight Claude Code model/provider switcher | 一个精简的 Claude Code 模型/服务商快速切换神器

Metadata-only control plane for request-driven switching across long-lived vLLM Switch backends

vLLM Switch: a research fork for exact CPU/disk weight backup, host-memory reclaim, and fast model switching

A reusable build framework + GUI app that manages AI coding agent configs (OpenCode, KiloCode). Split messy JSON into profiles, providers, models, plugins and MCP — rebuild with one click. Free, local-first, no cloud. Built by a learner, for learners.

🕹️ Downshift - Start premium. Finish cheap. Downshift cuts long Pi Coding Agent session costs by switching from your premium model to an economy model when context gets expensive.

A hybrid CNN-Transformer deep learning system for distinguishing COVID-19, Normal, and Pneumonia from chest X-ray images

Hermes Agent skill for dynamic AI model switching via AI API Proxy