ChatGPT Plus
$20/mo
Standard Codex access
Codex CLI
OpenAI's agentic coding mode — in your terminal, IDE, app, or pocket.
OpenAI's agentic coding product. The Codex CLI is open source, but Codex now spans a CLI, VS Code and JetBrains extensions, a desktop app, and remote control from the ChatGPT mobile app. Sign in with a ChatGPT plan or BYO API key. Models include GPT-5.5, GPT-5.4 family, and GPT-5.3-Codex.
+ Pros
- Multi-surface: CLI, VS Code/JetBrains extensions, desktop app, and ChatGPT mobile remote
- Tightly integrated with ChatGPT plans — no extra subscription for existing OpenAI users
- Run-iterate loop with sandboxed Docker execution ensures safe autonomous coding
- Plugin marketplace, Hooks, and Auto-review for flexible workflow automation
- Open-source CLI (npm) with BYO API key support
− Cons
- Ecosystem lock-in to OpenAI models — no support for Anthropic, Google, or local models
- Pro plan costs are high ($100–$200/mo) for heavy agentic use
- Docker sandbox required for full agent capabilities — adds setup complexity
- Younger multi-surface product — still maturing across IDE and mobile surfaces
Pricing
Pro (5x)
$100/mo
5x usage limits
Pro (20x)
$200/mo
20x usage limits
Business/Enterprise
Custom
Team and enterprise plans
API Key
Pay-per-token
BYO API key, standard pricing
Codex is OpenAI's agentic coding product. The CLI started as an open-source terminal tool that wrote code, ran it in a sandbox, and iterated until tasks were done. Over 2025–2026 it expanded into a full surface: the Codex CLI (npm, open source), an IDE extension (VS Code, JetBrains), the Codex app for desktop, and remote control from the ChatGPT mobile app via a connected host.
Codex is tightly integrated with ChatGPT plans — sign in with a ChatGPT account and your Plus, Pro, Business, or Enterprise plan covers Codex usage, with limits priced in credits per million tokens. Models include GPT-5.5 (recommended), GPT-5.4, GPT-5.4-mini, GPT-5.3-Codex, and GPT-5.3-Codex-Spark (research preview). The CLI and app run code in a Docker sandbox with configurable permission profiles, support a plugin marketplace, Hooks, Auto-review, and an in-app browser for verifying local UIs.
Key Features
• Multi-surface: CLI (open source), IDE extension (VS Code, JetBrains), Codex app
• ChatGPT mobile: drive Codex on a connected Mac host from your phone
• Models: GPT-5.5, GPT-5.4, GPT-5.4-mini, GPT-5.3-Codex, Codex-Spark
• Run-iterate loop: writes, runs in a sandbox, reads output, refines
• Plugin marketplace, Hooks, Auto-review for agent approvals
• In-app browser plugin for visual debugging of local dev servers
• BYO API key supported — falls back to standard API pricing
• ChatGPT plan auth (Plus, Pro $20/$100/$200, Business, Enterprise)
Version History
Added search across local conversation history with case-insensitive content matching. Made --profile the primary profile selector across CLI, TUI permissions.
Goals enabled by default with dedicated storage and progress tracking. codex remote-control now runs as a foreground command.
Python SDK first-class authentication (API key login, ChatGPT browser/device-code flows). codex exec resume now accepts --output-schema.
Richer TUI session controls with data-driven service-tier commands. @ mentions now search files, directories, plugins, and skills in one picker.
Plugin details now show bundled hooks, plugin sharing exposes link metadata. Added codex remote-control as simpler entrypoint for headless app-server.
Signature Snippet
$ codex 'Build a CLI tool that converts markdown to PDF and add tests'
⋮
Codex plans, scaffolds a Go module, writes the converter and tests, runs `go test`, captures the failures, and iterates until the test suite is green — all inside a sandboxed Docker workspace.More in this Space
Vix
Vix is a Go-native, open-source (AGPL-3.0) AI coding agent that slashes token costs by 40-50% using a stem agent architecture and Tree-sitter virtual filesystem. It rethinks the plan/execute loop — keeping LLM cache warm across Explore, Plan, and Execute phases — while shipping Programmable Workflows, Whiteboard Mode with voice AI, MCP server support, and a self-evolving agent that writes its own scheduled jobs and watchers.
Late — High-Leverage AI Agent Orchestration
Orchestrate an entire AI dev team on 5GB VRAM using ephemeral subagents, exact-match diffs, and a zero-dependency Go binary. Works with any OpenAI-compatible model — local or cloud.
Paca
Paca is a free, open-source, self-hosted Scrum board where AI agents work as equal teammates — assigned to sprints, picking up tasks, and collaborating on BDD specs alongside humans. Built as an alternative to Jira and Linear, it treats AI agents as first-class Scrum members.