How to use GPT, Gemini and other models in Claude Code in 2026: 8 ways compared

Disclosure: Enjoy is our product, and it runs Claude Code, Codex and OpenCode. We checked every way below against its maker’s own pages on November 9, 2026, and we say where another way is the better pick, including the free ones that don’t need Enjoy.

The ways to use other models with Claude Code, compared

Claude Code has two halves: the agent on your computer, which reads your project, edits files and runs commands, and the model behind it, which is Claude. Anthropic’s docs say it “doesn’t support routing Claude Code to non-Claude models through any gateway.” But Claude Code sends its requests to whatever address ANTHROPIC_BASE_URL names, and a lot of services now answer in Claude’s format. So there are three kinds of way: bring another lab’s agent into Claude Code, point Claude Code at another model, or keep each lab’s own agent and move the work between them.

Way	Best for	Price, checked November 9, 2026	Models	Runs on
OpenAI’s Codex plugin	GPT reviewing or taking over Claude’s work	Free; uses your ChatGPT plan, with the CLI listed from Plus at $20 a month, or an API key	OpenAI’s, through Codex	Wherever Claude Code and the Codex CLI run
OpenRouter	The most models through one key	Each model at its provider’s price, plus a 5.5% fee	600+ models from 80+ providers	Anywhere Claude Code runs
Claude Code Router	Different models for different requests, with fallbacks	Free and open source; you pay the providers	OpenAI, Gemini, DeepSeek, Kimi, Z.AI, OpenRouter and more	Mac, Windows, Linux
LiteLLM	Teams that want budgets and keys on their own server	Free and open source; you pay the providers	100+ providers	Your own computer or server
DeepSeek, GLM or Kimi, directly	A cheaper model with nothing in between	DeepSeek Flash $0.30 per million input tokens and $1.20 output, half off-peak; GLM Coding Plan from $18 a month; Kimi Plus $19 a month	That maker’s models	Anywhere Claude Code runs
Ollama	Free models on your own computer, or open models in Ollama’s cloud	Free locally; Pro $20 a month with $60 of cloud credits	Open models	Mac, Windows, Linux
OpenCode	An agent built to take any model	Free; OpenCode Go $10 a month	200+ providers, a ChatGPT plan or a local model	Mac, Windows, Linux
Enjoy	Each lab’s own agent in one conversation, on your own plans	A monthly subscription, free for 7 days, with your phone and one computer (pricing)	Claude Code, Codex, Antigravity, Grok Build, Muse Code, and OpenCode’s models	Mac, Windows, Linux

Prices are as each maker lists them in the US, before tax: per month for plans, per million tokens for APIs. Every way that points Claude Code somewhere else replaces your Claude login, so you pay that service instead of Anthropic.

What changes when Claude Code talks to another model

• Your Claude plan steps aside. While a gateway’s token is set, “requests carry that credential in place of a developer’s claude.ai subscription login, and the subscription’s usage limits don’t apply to them,” Anthropic says.
• Some features stop. “With the gateway configuration active, the environment selector doesn’t offer Anthropic-hosted cloud environments, and Remote Control is unavailable,” says Claude Code’s connection guide. Voice dictation needs the gateway’s token removed too.
• Claude Code doesn’t know the model. For a model name it doesn’t recognize, Claude Code sends “Thinking with a fixed budget rather than adaptive reasoning, and no effort or context management fields,” says its compatibility guide, and it “applies its own default context window,” LiteLLM notes. Set CLAUDE_CODE_AUTO_COMPACT_WINDOW to the model’s real window, as Kimi’s and LiteLLM’s guides do.
• It still calls home. Claude Code sends “nonessential background traffic outside the gateway path, to Anthropic and to third-party services such as GitHub,” such as version checks and telemetry, until you set CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC=1.

OpenAI’s Codex plugin, inside Claude Code

Best for: asking GPT to review Claude’s work, or handing it a task, without leaving Claude Code or paying by the token.

OpenAI’s plugin lets you “Use Codex from inside Claude Code for code reviews or to delegate tasks to Codex.” In Claude Code, run /plugin marketplace add openai/codex-plugin-cc, /plugin install codex@openai-codex, /reload-plugins and /codex:setup, which can install Codex for you. Then /codex:review is “a normal read-only Codex review,” /codex:adversarial-review is “a steerable challenge review,” /codex:rescue “Hands a task to Codex,” and /codex:transfer “Creates a persistent Codex thread from the current Claude Code session.” It’s open source under Apache 2.0, with about 35,000 stars on GitHub.

Price: free. It asks for a “ChatGPT subscription (incl. Free) or OpenAI API key,” and “Usage will contribute to your Codex usage limits.” OpenAI’s pricing lists Codex in the CLI from Plus, at $20 a month, and gives Free and Go “GPT-6 Luna at Standard speed in the desktop app, subject to rollout.”

Watch for: GPT works only on what you hand it; Claude still writes Claude Code’s own replies. The optional review gate “can create a long-running Claude/Codex loop and may drain usage limits quickly.” Its latest release, v1.0.6, is from July 8, 2026, and its examples still name older models such as gpt-5.4-mini. We found no plugin like it from Google for Gemini; community plugins wrap Google’s Antigravity command line, and we haven’t checked them.

OpenRouter

Best for: trying many labs’ models in Claude Code with one key and one bill.

OpenRouter answers in Claude’s format, so Claude Code needs no proxy. Its Claude Code guide sets ANTHROPIC_BASE_URL to https://openrouter.ai/api, your OpenRouter key as ANTHROPIC_AUTH_TOKEN, and ANTHROPIC_API_KEY to an empty string. Choose models with ANTHROPIC_DEFAULT_OPUS_MODEL, ANTHROPIC_DEFAULT_SONNET_MODEL and their siblings, or set CLAUDE_CODE_ENABLE_GATEWAY_MODEL_DISCOVERY=1 to pick from “a curated set of the highest-ranked models” in /model.

Price: pay as you go. “We pass through the pricing of the underlying model providers without any markup,” its FAQ says. It charges its fee when you buy credits: 5.5% on the Standard plan, says its pricing, for “600+ models” from “80+ providers.” The free plan has “25+ free models” at “50 requests/day.”

Watch for: its own guide opens with “Claude Code with OpenRouter is only guaranteed to work with the Anthropic first-party provider,” and adds “Claude Code is optimized for Anthropic models and may not work correctly with other providers.” Some entries in the model picker “route to non-Anthropic providers and may be subject to the compatibility limitations noted above.”

Claude Code Router

Best for: sending different requests to different models, with fallbacks when one fails, from a desktop app.

Claude Code Router “is a local model gateway and control plane for coding agents.” Its desktop app for macOS, Windows and Linux runs a gateway on your computer at http://127.0.0.1:3456. You add providers, apply a profile for Claude Code under Agent Config, and set rules under Routing for “conditions, retries, request rewrites, or fallback models.” It “supports OpenAI Chat / Responses, Anthropic Messages, Gemini Generate Content / Interactions, OpenRouter, DeepSeek, SiliconFlow, Moonshot, Kimi Code, Mistral, Z.AI, Bailian, and custom compatible providers,” and does the same for Codex, OpenCode and other agents. A command-line version, npm install -g @musistudio/claude-code-router, needs Node.js 22 or newer.

Price: free and open source under the MIT license, with about 38,000 stars on GitHub. You pay each provider.

Watch for: it only sees what’s launched through it. Its troubleshooting page says to “Confirm the agent was launched from CCR, not opened directly,” and “If any of them is wrong, requests bypass CCR.” Its latest release, v3.1.3, is from October 9, 2026.

LiteLLM

Best for: a team that wants everyone’s Claude Code on one gateway, with budgets and keys per person.

LiteLLM is a proxy you run yourself. Its tutorial installs it with uv tool install 'litellm[proxy]', lists your models in a config.yaml, starts it on port 4000 and points Claude Code at it with ANTHROPIC_BASE_URL and ANTHROPIC_AUTH_TOKEN. It “automatically translates between different provider formats,” across “100+ providers,” and its UI can “Set budget limits per user/key.”

Price: free and open source under the MIT license, apart from an enterprise/ folder with a license of its own. You pay the providers.

Watch for: it takes the most setup here. Its tutorial says Claude Code’s /model picker “only keeps gateway models whose id contains claude or anthropic,” so another model needs “a claude-flavored name” to appear in it. LiteLLM 1.107.0 carried a credential stealer for a few hours on October 21; its latest release is 1.108.3, from November 7.

DeepSeek, GLM or Kimi, directly

Best for: a cheaper model in Claude Code, with no gateway in between.

Several model makers answer in Claude’s format and publish the settings for Claude Code:

1. DeepSeek. Its guide sets ANTHROPIC_BASE_URL to https://api.deepseek.com/anthropic and Claude Code’s model variables to deepseek-flash. Its pricing is $0.30 per million input tokens and $1.20 per million output at peak, and “Off-peak rates are half of the peak rates.”
2. Z.ai’s GLM Coding Plan. Its Claude Code guide maps Opus, Sonnet and Haiku to GLM-5.3-Flash by default, and its plans start “at just 18 USD per month.”
3. Kimi Code. Kimi’s guide points Claude Code at https://api.kimi.ai/coding/ and sets every model variable to k3.1-pro. Kimi Code comes with Kimi’s memberships from Plus, at $15 a month billed yearly or $19 monthly; “The Go tier has no coding quota.” Our Kimi K3.1 Pro memo covers the model.

Price: as above, from each maker’s own pages.

Watch for: each has limits of its own. On Z.ai’s plans, “Each plan is subject to both a 5-hour usage limit and a weekly usage limit.” Kimi’s K3.1 Pro models “only support thinking mode,” and its guide warns that “protocol conversion through proxy tools may cause behavior to vary.” DeepSeek’s API maps “Models starting with claude-opus” to deepseek-v4-pro, “billed at the V4 Pro price,” so a setting left on an Opus name costs more. And its Flash model is the one an independent group found turning off its shutdown script in tests.

Ollama

Best for: models on your own computer, free, or open models in Ollama’s cloud.

“Ollama connects Claude Code to local and cloud models through its Anthropic-compatible API,” says its guide. Install Ollama for Mac, Windows or Linux, run ollama launch claude and pick a model. Our guide to local coding agents covers the memory and models it takes.

Price: free on your computer. Cloud models bill by the token, or come out of monthly credits: Pro is $20 a month with “$60 of usage credits per month,” and Max is $100 with “$300 of usage credits per month,” says its pricing.

Watch for: a local model needs room. Ollama’s guide says to “Choose a model and set a 64k+ context window,” but its default is 4k on graphics cards with under 24 GiB of memory. Its cloud models run on Ollama’s servers, not your computer.

OpenCode

Best for: an agent made to take any model, rather than Claude Code made to accept one.

OpenCode is “The open source coding agent”: “Start with free models or bring any of 200+ providers,” in a terminal or a desktop app for macOS, Windows and Linux. Run /connect to add a provider and /models to choose one, and it can “Use a ChatGPT Plus or Pro plan instead of an OpenAI API key,” where “Usage counts against your plan.” Our OpenCode vs Claude Code guide compares the two agents.

Price: free and open source under the MIT license. OpenCode Go, its subscription for “popular open coding models,” is $10 a month, and Go Plus is $40.

Watch for: it’s a different agent, with its own commands and settings, so you leave Claude Code’s behind. With a ChatGPT sign-in, OpenCode “caps the context window for these sign-ins below each model’s real limit,” to match Codex’s defaults.

Enjoy

Best for: each lab’s model in that lab’s own agent, on the plans you already pay for, in one conversation.

Enjoy doesn’t put another lab’s model inside Claude Code, and it doesn’t add one lab’s models to another lab’s account. It runs Claude Code, Codex, Antigravity, Grok Build, Muse Code and OpenCode on Mac, Windows and Linux, and one conversation can switch between them: “The new agent gets the conversation so far, including replies from other agents, before it works on your message.” So Claude Code can plan on your Claude plan, Codex can build on your ChatGPT plan, and Antigravity can review with your Google account, with no API key. For any other model, OpenCode in Enjoy uses the providers you set up in it, OpenRouter or a local model included. When a plan runs out, Enjoy switches to another of your accounts that still has usage, or waits for the limit to reset, then continues.

Price: one monthly subscription on top of your AI plans, and the first seven days are free, with a card to start. It includes your phone and any browser and one computer; each other computer costs extra each month, and each teammate you invite needs a subscription of their own. See pricing.

Watch for: Enjoy isn’t open source. Each agent signs in with its own lab’s account, so GPT here means Codex on a ChatGPT plan, not GPT inside Claude Code. Enjoy can’t show OpenCode’s usage, because it “depends on each provider.” Scheduled workflows run only while your computer is awake with Enjoy open.

When to stay with Claude in Claude Code

If Claude does the work well and your plan’s limits hold, stay. It’s the only model Anthropic supports in Claude Code, and Remote Control, voice dictation and Anthropic’s cloud environments don’t work through a gateway. A gateway also has to keep up, because “Claude Code adds capabilities with each release, and a gateway that doesn’t forward them breaks the corresponding features.” Anthropic’s pricing says “Claude Code is included in all paid plans,” from Pro at $17 a month billed yearly or $20 monthly.

For a second opinion, the Codex plugin adds GPT without touching Claude Code’s settings. And if you’re reaching for another model because you ran out, our guide to Claude Code’s usage limit compares the ways to keep going.

How we compared

We read Anthropic’s Claude Code docs on gateways, connecting to one, gateway compatibility and Remote Control; Claude’s and ChatGPT’s pricing; OpenAI’s Codex plugin README; OpenRouter’s Claude Code guide, pricing and FAQ; Claude Code Router’s README and troubleshooting page; LiteLLM’s tutorial and license; DeepSeek’s, Z.ai’s and Kimi’s Claude Code guides, plans and pricing; Ollama’s Claude Code guide, pricing and downloads; OpenCode’s homepage and docs; and Enjoy’s own help on November 9, 2026, and every one is linked below. Licenses, star counts and release dates come from GitHub that day. We quote each maker’s own words for its limits and leave out what it hasn’t documented, and we didn’t run these setups ourselves. These tools change weekly, so we recheck this guide against its sources at least once a month, and the date at the foot of the page says when we last did.