Claude Code Router is an open-source local model gateway for developers who use coding agents and want to manage their model providers in one place. It runs on macOS, Windows and Linux, with Docker and a CLI with a browser interface also available. The project uses the MIT license.
The gateway connects Claude Code, Codex, OpenCode, Claude Design and other supported agents to a shared local endpoint. You can switch models or providers without repeatedly changing each agent's configuration. Provider support includes OpenAI, Anthropic, Gemini, OpenRouter, DeepSeek, Moonshot and custom compatible endpoints. The gateway runs locally; requests routed to cloud providers go to those services for processing.
Routing rules can choose models based on request conditions, rewrite requests and retry failures. Ordered fallback models let requests move to another model when the first choice fails, while credential pools and key rotation handle multiple provider accounts. A searchable model catalog helps with model selection.
Fusion adds vision, web search and MCP tools to existing models. ToolHub, plugins and built-in browser automation extend the tools available to agents. Request logs show which provider, model and credential handled a call, alongside latency, token use, estimated cost and errors. The interface also exposes tool calls and agent traces. AgentClaw relays agents through messaging services including Slack, Discord, Telegram and Feishu.
Claim this page and we'll verify you by hand. Claude Code Router gets the verified badge, and you can upgrade the listing to be featured on localhosted. Proud to be listed? Put our badge on your site.
Want more people to find Claude Code Router?Promote it
Something wrong or outdated on this page?
1.2KUpdated 1 day agoMIT
macOS · Windows · Linux · Docker · Web#Guardrails#llama.cpp backend#LLM tracing
GoModel is a self-hosted AI gateway for developers and platform teams that want one API for local models and cloud providers. It accepts OpenAI- and Anthropic-compatible requests, so applications can keep their existing SDKs while the gateway handles provider selection and usage controls.
28.2KUpdated 1 day agoApache-2.0
Docker · Web#Batch processing#LLM tracing#MCP
11.7KUpdated 4 months agoApache-2.0
Docker · Web#Batch processing#LLM tracing#Multimodal input
9.5KUpdated 1 day agoApache-2.0
Docker · Web#Guardrails#MCP#Tool calling
2.2KUpdated 20 hours agoApache-2.0
Docker#LLM tracing#MCP#Multi-user access
8.5KUpdated 23 hours agoApache-2.0
Web#LLM tracing#MCP#Multimodal input
MLflow brings agent tracing, LLM evaluation, and model experiment tracking into a platform you can run locally or on your own servers. It's for developers and teams who need to understand failures, compare changes, and monitor AI applications in production. It's open source under Apache 2.0.
TensorZero is a self-hosted platform for developers building LLM applications. The project is archived and no longer maintained. It combines a model gateway with tools for inspecting responses, evaluating workflows, and improving prompts using production data and human feedback.
Higress is a self-hosted AI gateway for developers and teams managing model APIs and the tools their AI agents call. It puts LLM traffic and MCP servers behind a shared entry point, with authentication, traffic controls and monitoring. The open-source edition uses the Apache 2.0 license and runs locally in Docker without registration. Alibaba Cloud also offers a fully managed gateway.
Agent Router is an open source AI gateway for teams whose agents use both model APIs and MCP tools. It runs on a laptop, a dedicated gateway, or Kubernetes, and gives applications one OpenAI-compatible entry point for cloud providers and self-hosted inference. The project uses the Apache 2.0 license.
Bifrost is a self-hosted AI gateway for developers and teams whose applications use multiple model providers. It puts Ollama, custom model deployments, and cloud services behind one OpenAI-compatible API, so applications can switch models without maintaining a separate integration for each provider.