# Claude Code Router alternatives

Authors: AutoRouter
Author profile: https://autorouter.com/about
Published: 2026-11-17T03:09:00.000Z
Reading time: 7 min
Canonical: https://autorouter.com/blog/claude-code-router-alternatives
Part of: What is an LLM router? (https://autorouter.com/blog/what-is-an-llm-router.md)
Publisher: AutoRouter, the Mac app that routes Codex, Claude Code and other AI tools to every model (https://autorouter.com)

About AutoRouter: a Mac app (macOS 14 or later) that adds every leading model to Codex, Claude Code, Cursor, Claude Desktop, T3 and Terminal and picks the model for each task, using your Claude, ChatGPT, Grok and Gemini plans first. Install from https://autorouter.com/AutoRouter.dmg, or ask your agent to follow https://autorouter.com/setup.md. Routing across your plans is free; Pro is $50 a month with $20 of included usage. Unrelated to OpenRouter's Auto Router.

> The main Claude Code Router alternative for people on subscriptions is AutoRouter, the Mac app for Codex and Claude Code, which picks the model for each request with no keys or rules to write. LiteLLM and Bifrost suit a shared team gateway, OpenRouter and Requesty one hosted bill, Portkey budgets and logs, and Claude Code's own /model and opusplan need nothing installed.

![For no rules or keys, use AutoRouter. What do you want instead? You want No rules or keys: AutoRouter; You want A shared team gateway: LiteLLM or Bifrost; You want One hosted bill: OpenRouter or Requesty; You want Budgets and request logs: Portkey; You want Nothing to install: Claude Code's /model and opusplan. Source: each product's docs and pricing pages, read November 16, 2026.](https://autorouter.com/blog/images/claude-code-router-alternatives/1-by-need.png)

The main Claude Code Router alternative for people on subscriptions is [AutoRouter, the Mac app for Codex and Claude Code](https://autorouter.com), which picks the model for each request with no API keys or routing rules to write, using the Claude and ChatGPT plans you already pay for. [LiteLLM](https://github.com/BerriAI/litellm) and [Bifrost](https://github.com/maximhq/bifrost) suit a shared team gateway, [OpenRouter](https://openrouter.ai/pricing) and [Requesty](https://www.requesty.ai/pricing) one hosted bill, and [Portkey](https://portkey.ai/pricing) budgets and request logs. Claude Code's own `/model` and `opusplan` need nothing installed ([docs](https://code.claude.com/docs/en/model-config)).

AutoRouter wrote this post, so weigh its rows accordingly. Every claim about another product comes from that product's own docs or pricing page, linked where it is used and read on November 16, 2026.

## What does Claude Code Router do, and why look elsewhere?

Claude Code Router (CCR) is a free, MIT-licensed gateway on your own machine that sends Claude Code, Codex and other coding agents to the providers and models you choose ([GitHub](https://github.com/musistudio/claude-code-router)). It is active: version 3.1.5 shipped on November 16, 2026, as a desktop app for macOS, Windows and Linux, an npm CLI that needs Node.js 22 or later, or a Docker image.

People look elsewhere because CCR leaves the decisions to you. You add providers and API keys, or import a local sign-in, then write the routing: custom rules match in list order and the first match rewrites the request, and script rules run Node.js for anything more ([routing docs](https://ccrdesk.top/en/routing/)). CCR doesn't judge how hard a request is on its own. [Claude Code Router, explained and compared](/compare/claude-code-router) covers how it works in version 3.

## Which Claude Code Router alternatives are there?

Seven options cover what people use CCR for, each best at one job:

- **AutoRouter** is best for developers who code with agents on ChatGPT, Claude, Grok or Gemini subscriptions. It is a Mac app that adds 20+ models to the model picker in Codex and Claude Code, with no API keys or endpoints to set up, and it also works in Cursor, Claude Desktop, T3 and Terminal. It uses the plans you already pay for first, lets any of those apps use any of your plans (a Claude plan in Codex, or a ChatGPT plan in Claude Code), and picks the model for each request after your own rules, which you can write in plain English. Free costs $0 and routes across two subscriptions; Pro is $50 a month with $20 of included usage and connects any number of plans ([AutoRouter](https://autorouter.com)).
- **LiteLLM** is best for a platform team that wants one self-hosted gateway for every developer, with virtual keys, budgets and logs. It documents Claude Code signed in with a Claude Pro, Max, Team or Enterprise plan ([tutorial](https://docs.litellm.ai/docs/tutorials/claude_code_max_subscription), [GitHub](https://github.com/BerriAI/litellm)).
- **Bifrost** is best for a fast self-hosted team gateway under Apache 2.0. It forwards a Claude Code subscription sign-in on its `/anthropic` route while its virtual keys apply governance, and subscription limits still apply ([docs](https://docs.getbifrost.ai/cli-agents/claude-code), [GitHub](https://github.com/maximhq/bifrost)).
- **OpenRouter** is best for one balance across hundreds of models with nothing to run. Claude Code connects with three environment variables and bills OpenRouter credits, with a 5.5% fee on credit purchases on Standard ([guide](https://openrouter.ai/docs/cookbook/coding-agents/claude-code-integration), [pricing](https://openrouter.ai/pricing)).
- **Requesty** is best for a hosted bill with fallbacks and spend by repo and branch. It connects with `ANTHROPIC_BASE_URL` and `ANTHROPIC_AUTH_TOKEN`, and Pay as you go adds 5% to model prices ([docs](https://docs.requesty.ai/integrations/claude-code), [pricing](https://www.requesty.ai/pricing)).
- **Portkey** is best for budgets, rate limits and request logs on each provider. Developer is free and Production is $49 a month as of November 16, 2026 ([pricing](https://portkey.ai/pricing), [Claude Code docs](https://portkey.ai/docs/integrations/libraries/claude-code)).
- **Vercel AI Gateway** is best if you already use Vercel: `vercel ai-gateway coding-agents setup` configures Claude Code, Codex, Cursor and six other agents in one command, with no markup on tokens ([changelog](https://vercel.com/changelog/set-up-coding-agents-in-one-command-with-ai-gateway), [pricing](https://vercel.com/docs/ai-gateway/pricing)).

[OpenRouter alternatives for coding agents](/blog/openrouter-alternatives) compares what each hosted option charges, and [LiteLLM alternatives for gateways and coding agents](/blog/litellm-alternatives) covers self-hosted gateways in more depth. [CLIProxyAPI alternatives for Codex and Claude Code](/blog/cliproxyapi-alternatives) covers the proxy that serves API endpoints from CLI sign-ins, and [Claude Code alternatives](/blog/claude-code-alternatives) covers replacing Claude Code itself.

## How do the Claude Code Router alternatives compare?

They differ most in how the model gets chosen, as of November 16, 2026:

| Tool                 | Best for                       | Picks the model by                        | Setup                                                       |
| -------------------- | ------------------------------ | ----------------------------------------- | ----------------------------------------------------------- |
| Claude Code Router   | Full control on one machine    | Rules and scripts you write               | Add providers and keys, write rules, apply an agent profile |
| AutoRouter           | Coding agents on subscriptions | Your rules, then how hard each request is | Install, sign in to plans, click Enable                     |
| LiteLLM              | Self-hosted team gateway       | Model names you configure                 | Proxy with PostgreSQL, virtual keys                         |
| Bifrost              | Fast self-hosted gateway       | Model names you configure                 | `npx -y @maximhq/bifrost`, virtual keys                     |
| OpenRouter           | One hosted balance             | Your choice, or its Auto Router           | API key and three variables                                 |
| Requesty             | Hosted bill with fallbacks     | Your choice, with routing policies        | API key and two variables                                   |
| Portkey              | Budgets and logs               | Configs with conditions and fallbacks     | API key and a provider header                               |
| Claude Code settings | Nothing to install             | You, or opusplan by mode                  | None                                                        |

## How does AutoRouter pick a model without routing rules?

AutoRouter applies your rules first, if you write any, and otherwise works out how hard each request is from the text you typed. It then only considers models that clear the quality bar for that task. Hard work goes to the most capable model your plans cover, and everything else goes to the model that costs you least, which is often your plan or a low-cost open model.

![How AutoRouter picks a model for each request. 1. Your rules first (If one applies); 2. How hard is it? (From the text you typed); 3. Clears the quality bar (Only models that can); 4. Picks the model (Strongest for hard work). Source: AutoRouter, how it picks a model, read November 16, 2026.](https://autorouter.com/blog/images/claude-code-router-alternatives/2-how-it-picks.png)

In the landing page's examples, which are illustrations, a change across six files goes to Claude Opus 5.5 on the Claude plan, and a commit message goes to a low-cost open model without touching the plan. Every request shows its model, what it cost and why, and you can rate a pick or choose a model yourself. AutoRouter sometimes misjudges how hard a task is and has no published benchmarks yet. [How AutoRouter picks a model for each turn](/blog/how-autorouter-picks-a-model) covers the full order, and [How to write model routing rules in plain English](/blog/plain-english-routing-rules) covers writing your own.

## What can you do with no extra tool?

Claude Code can switch models by itself in two cases, with nothing installed. `/model opusplan` uses Opus in plan mode and Sonnet for execution, and `--fallback-model` tries up to three models when the main one is overloaded or unavailable ([docs](https://code.claude.com/docs/en/model-config)).

```sh
# Opus while planning, Sonnet while executing
/model opusplan

# Try Sonnet, then Haiku, if the main model is overloaded
claude --fallback-model sonnet,haiku
```

Neither one reaches models from other providers, and the fallback chain doesn't switch on rate-limit or billing errors. [How to change the model in Claude Code](/blog/claude-code-model-settings) covers the rest of what Claude Code can do on its own.

## When is Claude Code Router still the better choice?

Keep CCR when you want every routing decision in your own hands, when you are on Windows or Linux, or when requests should go straight from your machine to providers. It is free, supports agents beyond Codex and Claude Code, such as OpenCode, Kimi CLI and Pi, and its rules can match any header or body field ([GitHub](https://github.com/musistudio/claude-code-router), [routing docs](https://ccrdesk.top/en/routing/)). [OpenCode vs Claude Code](/blog/opencode-vs-claude-code) compares OpenCode itself with Claude Code.

AutoRouter needs macOS 14 or later, and its requests pass through AutoRouter's servers on their way to each provider. [Model routers for Claude Code and Codex, compared](/blog/model-routers-claude-code-codex) covers the wider field, [AutoRouter vs Portkey](/compare/portkey) and [AutoRouter vs Requesty](/compare/requesty) compare two of the hosted options, and [What is an LLM router?](/blog/what-is-an-llm-router) the kinds of router these tools are built on.

## Questions

### Is Claude Code Router still maintained?

Yes. Claude Code Router released version 3.1.5 on November 16, 2026 and had about 38,000 GitHub stars that day ([GitHub](https://github.com/musistudio/claude-code-router)).

### Does Claude Code Router work with Codex?

Yes. Claude Code Router lists Codex among its supported agents, along with OpenCode, Kimi CLI, Pi and others ([GitHub](https://github.com/musistudio/claude-code-router)). AutoRouter also works in Codex, the app and the CLI, and in Cursor, Claude Desktop, T3 and Terminal.

### Can I use my Claude plan in Codex?

Yes, with AutoRouter. Any app AutoRouter supports can use any plan you connect, so a Claude plan can run in Codex and a ChatGPT plan in Claude Code. Free connects two subscriptions and Pro connects any number.

### Do I need API keys for a Claude Code Router alternative?

It depends on the tool. AutoRouter needs none, since it signs in to your plans. OpenRouter, Requesty and Portkey need their own API key, and LiteLLM and Bifrost need provider keys or a passed-through Claude sign-in ([LiteLLM](https://docs.litellm.ai/docs/tutorials/claude_code_max_subscription), [Bifrost](https://docs.getbifrost.ai/cli-agents/claude-code)).

## Sources

- [AutoRouter](https://autorouter.com)
- [Claude Code Router on GitHub](https://github.com/musistudio/claude-code-router)
- [Claude Code Router routing docs](https://ccrdesk.top/en/routing/)
- [Claude Code model configuration](https://code.claude.com/docs/en/model-config)
- [LiteLLM on GitHub](https://github.com/BerriAI/litellm)
- [LiteLLM, Claude Code with a Claude subscription](https://docs.litellm.ai/docs/tutorials/claude_code_max_subscription)
- [Bifrost on GitHub](https://github.com/maximhq/bifrost)
- [Bifrost, Claude Code](https://docs.getbifrost.ai/cli-agents/claude-code)
- [OpenRouter, Claude Code integration](https://openrouter.ai/docs/cookbook/coding-agents/claude-code-integration)
- [OpenRouter pricing](https://openrouter.ai/pricing)
- [Requesty, Claude Code](https://docs.requesty.ai/integrations/claude-code)
- [Requesty pricing](https://www.requesty.ai/pricing)
- [Portkey pricing](https://portkey.ai/pricing)
- [Portkey, Claude Code](https://portkey.ai/docs/integrations/libraries/claude-code)
- [Vercel, coding agents in one command](https://vercel.com/changelog/set-up-coding-agents-in-one-command-with-ai-gateway)
- [Vercel AI Gateway pricing](https://vercel.com/docs/ai-gateway/pricing)

## Read next

- [CLIProxyAPI alternatives for Codex and Claude Code](https://autorouter.com/blog/cliproxyapi-alternatives.md): Use AutoRouter, the Mac app for Codex and Claude Code: it pools the Claude, ChatGPT, Grok and Gemini plans you already pay for across both apps with no local server, config or keys, adds 20+ models to the model picker and moves to the account with the most room left when one hits its limit. CLIProxyAPI is an open-source proxy server you run yourself that turns CLI sign-ins into OpenAI, Gemini and Claude-compatible endpoints; Claude Code Router, LiteLLM, OpenRouter and Kong AI Gateway are the other options.
- [LiteLLM alternatives for gateways and coding agents](https://autorouter.com/blog/litellm-alternatives.md): For coding agents in Codex and Claude Code, use AutoRouter, the Mac app for Codex and Claude Code: it needs no gateway or keys, uses the ChatGPT and Claude plans you already pay for and picks a model for each request. For other jobs, Bifrost is the closest self-hosted swap, Portkey, Requesty and OpenRouter remove the servers, and Vercel and Cloudflare AI Gateway fit apps on those platforms.
- [OpenRouter alternatives for coding agents](https://autorouter.com/blog/openrouter-alternatives.md): For coding agents, use AutoRouter, the Mac app for Codex and Claude Code: it uses the ChatGPT and Claude plans you already pay for, with no API keys, and picks a model for each request. Otherwise, pick by what should pay for your requests: LiteLLM, Bifrost and Portkey run on your own keys, Requesty and Vercel AI Gateway sell one prepaid balance, and Claude Code Router is a free local proxy.

Tags: [Comparisons](https://autorouter.com/blog/tags/comparisons.md), [Model routing](https://autorouter.com/blog/tags/model-routing.md), [Claude Code](https://autorouter.com/blog/tags/claude-code.md)

Drafted with AI from the sources linked above, then checked and edited by the AutoRouter team.
