Comparison

Codex vs Gemini CLI

How the two compare on the things that decide whether your setup carries over: open standards, license, models and where they run.

CodexGemini CLI
MakerOpenAIGoogle
LicenseApache-2.0Apache-2.0
ModelsOpenAI modelsGoogle Gemini models
Runs inTerminal, IDE, Desktop, WebTerminal
MCPMCP: supportedMCP: supported
Agent SkillsAgent Skills: supportedAgent Skills: supported
AGENTS.mdAGENTS.md: supportedAGENTS.md: supported
ACPACP: supportedACP: supported

Key differences

  • Both are open source (Apache-2.0 and Apache-2.0), so you can inspect and self-host either.
  • Models: Codex runs OpenAI models; Gemini CLI runs Google Gemini models.
  • Codex also runs in IDE, Desktop, Web.
  • Both support MCP, Agent Skills, AGENTS.md, ACP, so MCP servers, skills and repo instructions you set up carry over if you switch.

In their own words

Codex

OpenAI's coding agent across CLI, IDE, ChatGPT and cloud. One of the originators of AGENTS.md, with Skills, MCP and a plugin marketplace.

Gemini CLI

Google's open-source terminal agent for Gemini models, with MCP servers, extensions, Skills and AGENTS.md support.

FAQ

Is Codex or Gemini CLI open source?
Codex is open source under Apache-2.0. Gemini CLI is open source under Apache-2.0.
Do Codex and Gemini CLI support MCP servers?
Codex: yes. Gemini CLI: yes.
Can I reuse Agent Skills between Codex and Gemini CLI?
Yes. Both read the open Agent Skills (SKILL.md) format, so the same skill folder works in both.

Set up MCP servers in either one

More comparisons

Reviewed Oct 5, 2026.