HolyGrAIL

All tasks / Tools

Codex

OpenAI's coding agent for the terminal, IDE, desktop app, web and iOS; the CLI is open source.

  • OpenAI
  • Cloud
  • Free tier · from $8/mo
  • Open source

Official site · Source code

Where it ranks

  • A
    Coding
    Coding agents · Cloud · 63 on Artificial Analysis Coding Agent Index (v1.5)

Ranked by its own benchmark result where it has one; otherwise by the best model it offers.

Facts

Vendor
OpenAI
Approach
Coding agents
Where it runs
Cloud only: needs the vendor’s service
Interface
Command line, IDE extension, Desktop app, Web app, Mobile app
Skill level
Intermediate: some setup or configuration
Price
Free tier · from $8/mo
Pricing notes
Free: GPT-6 Luna in the desktop app (subject to rollout). Go $8/mo, Plus $20/mo, Pro $100/$200/$500/mo, Business $25/user/mo ($20 billed annually); or API key at API rates.
Source code
Open source (Apache-2.0)
GitHub stars
129.7k (18 Oct 2026)
Last verified
18 Oct 2026

Price from developers.openai.com, 18 Oct 2026.

Models it gives you

Its own benchmark results

BenchmarkResultEntry on the leaderboardAs ofSource
Artificial Analysis Coding Agent Index (v1.5)63Codex + GPT-6.1 Sol (xhigh)
AA label 'Codex - GPT-6.1 Sol (xhigh) ({'reasoning_effort': 'xhigh'})'; Coding Agent Index v1.5, rank 7 of 42; components (pass@1 avg of 3 attempts): DeepSWE v1.1 73.2%; SWE-Atlas-QnA 61.0%; Terminal-Bench v4 54.5%; read from the page's embedded data; no last-updated date shown
18 Oct 2026artificialanalysis.ai
Terminal-Bench 2.082.2%Codex CLI + GPT-5.5
rank 4 of 142 entries; raw accuracy 82.25 ± 2.22 (95% CI half-width); entry date 2026-04-23; marked verified; agent org: OpenAI; read via the site's leaderboard-read API (page is client-rendered)
18 Oct 2026tbench.ai
Terminal-Bench 2.187.4%Codex + GPT-6 Astra
reasoning effort: high; rank 1 of 22 entries; raw accuracy 87.42 ± 1.79 (95% CI half-width); entry date 2026-09-03; agent org: OpenAI; read via the site's leaderboard-read API (page is client-rendered)
18 Oct 2026tbench.ai
Terminal-Bench 3.034.6%Codex + GPT-5.6 Sol
reasoning effort: max; rank 2 of 12 entries; raw accuracy 34.59 ± 3.1 (95% CI half-width); entry date 2026-07-13; agent org: OpenAI; read via the site's leaderboard-read API (page is client-rendered)
18 Oct 2026tbench.ai
Terminal-Bench 4.058.2%Codex + GPT-6 Astra
reasoning effort: max; rank 3 of 35 entries; raw accuracy 58.18 ± 2.79 (95% CI half-width); entry date 2026-09-03; agent org: OpenAI; tied with GPT-6.1 Sol (max); read via the site's leaderboard-read API (page is client-rendered)
18 Oct 2026tbench.ai

Compare

Sources