↓ Skip to main content
tomrochette.com
  1. Agents/
  2. Harnesses/

Codex

Author
glm-5.3, x-preview-f-free, glm-5.3-flash
History / Edit / PDF / EPUB / MD
Created: / Updated: / Status: finished / Readability: 3 / 8 minutes read (~1555 words)
agent-curated 610 coding-agents 52 developer-tools 33 fully-ai-generated 715 harnesses 36 llm=glm-5.3 169 llm=glm-5.3-flash 620 llm=x-preview-f-free 22 openai 6 research-note 580
Table of Contents

Codex is OpenAI’s coding agent: a Rust CLI, an IDE extension, a desktop app, and a cloud service, all included in ChatGPT plans.

Codex is the only big-lab harness whose Apache-2.0 CLI an individual can still just run, and that matters more day to day than any single feature difference with Claude Code.

What it is
#

The codex CLI runs locally against your repository, with /init writing an AGENTS.md, sandboxed permissions, /review for local code review, skills and a plugin marketplace, subagents, and codex exec for scripts and CI. Codex cloud runs the same agent in cloud environments reachable from web, CLI, and GitHub (@codex on issues and PRs), and the SDK, app server, and MCP server expose the harness as a platform. Default models are the GPT-6 family (Astra, Sol, and Luna) rolling out across plans alongside the GPT-5.6 family (Sol, Terra, Luna), with GPT-6.1 Sol now listed on Plus and Pro for Work and Codex and GPT-5.5 retiring from ChatGPT on October 14, 2026.

Status
#

Active. openai/codex shows about 129.3k stars, about 20.3k forks, and about 12.3k commits under Apache-2.0 as of 2026-10-15.

Star History Chart

The CLI launched April 2025 and has been rewritten and rebuilt since; the current docs position it as an open agent harness platform. The release train stays hot: rust-v0.168.0 (November 13, 2026) is the latest stable, adding a Sessions view for browsing model providers and choosing the default for new chats, an opt-in features.tui_workspace side conversation beside the main chat with a resizable divider and leader z zoom, audio-recording attachments for audio-capable models (confirmed-rejected recordings skipped on later turns, clip duration used to keep short recordings in context), an opt-in features.image_generation_animation for generating and editing animated GIFs with frame count and timing controls, fullscreen image previews and pets in tmux on Ghostty and Kitty with allow-passthrough all, and tui.notification_method = "osc777" terminal notifications with longer command previews, after rust-v0.167.1 (November 10, 2026), a patch cut with no hand-written notes, after rust-v0.167.0 (November 6, 2026), adding codex agents --no-daemon to open the agent command center without the daemon, unread-result markers in the command center and session picker, backend switching from the terminal session picker that keeps drafts, local image attachment previews while composing with clickable transcript paths opening viewed images, API key and cloud provider options under “Sign in another way” with a gcp_setup_wizard for Vertex AI onboarding, opt-in macOS and Windows system proxy settings, and patch details expanding in the transcript beside an open approval prompt, after rust-v0.166.0 (October 29, 2026) added keyboard-driven transcript selection with word, line, and page motions and integrated search, consistently rendered Markdown in async questions, choices, and replies, GPT-6.1 Terra in the bundled, Vertex AI, and Amazon Bedrock catalogs, codex queue --resume for resuming unloaded queued threads, clearer safety-block explanations, and a codex doctor --network-only diagnostic mode, after rust-v0.165.0 (October 27) added a built-in Google Cloud Vertex AI provider configured with model_provider = "gcp", GPT-6.1 Terra in the model catalogs, a repository filter in the session picker with directory grouping in the agent command center, experimental attached thread forks in the app-server API, and mid-conversation reasoning-effort changes enabled by default, and rust-v0.164.0 (October 23) added ten numbered /presets slots for reusable prompts and slash commands with a keyboard menu, inline /model and /permissions arguments, and image previews in the fullscreen transcript on Kitty and Sixel terminals.

Strengths
#

  • Open-source CLI you can audit, fork, and drive with an API key, no subscription required.
  • Included from the ChatGPT Free tier upward, so the entry price is zero.
  • AGENTS.md is the first-class configuration convention, not a vendor variant.
  • Model tiering (Sol/Terra/Luna) makes high-volume work cheaper without changing tools.
  • Sandboxing and permission profiles are documented and prominent.

Cautions
#

  • API-key mode drops the cloud features (GitHub code review, Slack integration); those need a ChatGPT plan.
  • Usage is metered in shared five-hour windows across ChatGPT and Codex features, and the published per-window message ranges are wide (for example 10-100 local Sol messages on Plus).
  • The April 2025 launch drew sharp comparisons: early builds hallucinated docs Claude Code wrote correctly, and the original TypeScript CLI was rough; judge current releases, not that launch thread.
  • Skills and plugins are converging on formats every vendor now claims, so expect churn.

Pricing
#

Included in ChatGPT Free, Go ($8/month), Plus ($20/month), Pro (three tiers at $100, $200, and $500/month, the $500 tier carrying Ultrafast Astra access), Business ($20-25 per user), Enterprise and Edu. Overage is sold as credits; API-key usage is plain token pricing.

Price history
#

Date Plan Change Source
2026-09 ChatGPT plans Baseline: included from Free up, Go $8/mo, Plus $20/mo, Pro 5x $100/mo, Pro 20x $200/mo, Business $20-25 per user, Enterprise and Edu custom; overage sold as credits. learn.chatgpt.com/docs/pricing
2026-10-03 Pro The pricing page now lists Pro as three tiers at $100, $200, and $500 per month, where the note previously carried Pro 5x $100 and Pro 20x $200; the $500 tier is listed with Ultrafast Astra access. learn.chatgpt.com/docs/pricing

Compared to
#

  • Claude Code: deeper orchestration layer, heavier token baseline, closed client; pick Anthropic’s stack only if you are all-in on Claude models.
  • Gemini CLI: also Apache-2.0, but Google ended free consumer access in June 2026.
  • OpenCode: provider-neutral and MIT, but no bundled subscription usage.

Bottom line
#

Recommended for anyone already paying for ChatGPT: it is the subscription-backed harness with the least lock-in. My own split: work that suits a subscription goes here, metered work goes to aider or OpenCode on my keys. Not the cheapest path if your usage is light and token-based (aider or OpenCode with your own keys costs less).

Changes
#

  • 2026-08-22 - Created in the Harnesses category seed.
  • 2026-08-23 - Bolded each section’s key insight, named the launch-thread referent in the April 2025 caution, and set the updated field.
  • 2026-08-26 - Qualified the overbroad only-open-source-CLI thesis to what distinguishes it from Gemini CLI.
  • 2026-08-27 - Moved the three doc references to learn.chatgpt.com after the domain migration and reworded a banned-term phrase.
  • 2026-09-16 - Refreshed repository scale to about 125k stars, about 19.3k forks, and about 10.8k commits.
  • 2026-09-20 - Added the Price history section tracking price changes in a table, per the new owner rule.
  • 2026-09-22 - Recorded the GPT-6 family (Astra, Sol, Luna) arriving across ChatGPT plans alongside GPT-5.6 and the October 14, 2026 GPT-5.5 retirement, and refreshed repository scale.
  • 2026-09-26 - Linked the ChatGPT plans note in the new Model access category, where the subscription that carries this harness is tracked with its price history.
  • 2026-10-03 - Recorded the pricing page listing Pro as three tiers at $100, $200, and $500 per month (previously Pro 5x $100 and Pro 20x $200), with the $500 tier carrying Ultrafast Astra access, and appended the matching Price history row.
  • 2026-10-07 - Added the openai/codex star history chart to the Status section.
  • 2026-10-08 - Recorded the pricing page listing GPT-6.1 Sol on Plus and Pro for Work and Codex (its credit rate card added alongside), with the GPT-5.5 October 14 retirement reconfirmed.
  • 2026-10-17 - Recorded the stable train reaching rust-v0.163.0 (October 15), which added the searchable /config, follow-up steering of a responding turn with Code Mode cell interruption, and a configurable leader key, with the v0.164.0 alphas following, and re-verified the ChatGPT pricing page unchanged (Pro still $100/$200/$500 with Ultrafast on the $500 tier).
  • 2026-10-25 - Rotation re-verification, every cited reference live, and recorded rust-v0.164.0 as the latest stable (October 23): ten numbered /presets slots, inline /model and /permissions arguments, and Kitty/Sixel image previews in the fullscreen transcript.
  • 2026-10-27 - Recorded the v0.165.0 alpha train running (three alphas published October 26 through 27) behind the unchanged rust-v0.164.0 stable.
  • 2026-10-28 - Recorded the stable channel promoting rust-v0.165.0 (October 27): the built-in Google Cloud Vertex AI provider, GPT-6.1 Terra in the catalogs, session-picker repository filtering with agent-command-center directory grouping, experimental attached thread forks, and reasoning-effort switching mid-conversation, with the v0.166.0 alpha train already running.
  • 2026-11-01 - Recorded the stable channel promoting rust-v0.166.0 (October 29): keyboard-driven transcript selection with integrated search, rendered Markdown in Q&A surfaces, GPT-6.1 Terra across the bundled, Vertex AI, and Bedrock catalogs, codex queue --resume, clearer safety-block explanations, and codex doctor --network-only.
  • 2026-11-06 - Recorded rust-v0.167.0 as the latest stable (November 6): codex agents --no-daemon, unread-result markers, draft-keeping backend switching, image attachment previews while composing, Vertex AI setup in TUI onboarding, and opt-in system proxy settings.
  • 2026-11-13 - Recorded rust-v0.168.0 as the latest stable (November 13): a Sessions view for browsing model providers and choosing defaults, an opt-in TUI workspace side conversation, audio-recording attachments, opt-in animated GIF generation, tmux passthrough image previews on Ghostty and Kitty, and OSC 777 notifications.

See also
#

References
#