How Oracle turns days of work into minutes with ChatGPT and Codex (original report, opens in a new tab)
Read the original report from OpenAI.
Automatically collected from the publisher’s feed. Open the headline to read the original report.
Continue your Atlas path← Code
Initial research 2026-09-21OpenAI
Codex
Inspect the documented capabilities, access conditions and sources for your task.
Initial research: 2026-09-21. Check current terms before choosing.
Updated
The facts
- Version
- CLI rust-v0.154.0 (mid-September 2026)
- Price
- Price not verified. Open the tool record for the available pricing sources.
- API
✓Supported - Open weights
✕Not supported - Commercial use
- yes
- Agentic editing
✓Supported Multi-file agentic editing; the Codex harness is now exposed directly through the Agents API
Source for Agentic editing (opens in a new tab)- Terminal use
✓Supported CLI runs locally with configurable approval modes; 58.2% on Terminal-Bench 4.0 driving GPT-6 Astra — the top score, inside the margin of error against Claude Code
Source for Terminal use (opens in a new tab)- I D E integration
✓Supported Available in the terminal, the IDE and the browser
Source for I D E integration (opens in a new tab)- Autonomous P Rs
✓Supported Listed among the agents that take an issue through to an opened pull request
Source for Autonomous P Rs (opens in a new tab)
Sources
RELATED AI NEWS
LegalOn halves Codex costs while maintaining development speed (original report, opens in a new tab)
Read the original report from OpenAI.
Automatically collected from the publisher’s feed. Open the headline to read the original report.
Plugin4Shell: Zero-Click RCE Hits Claude Code, Codex, Copilot and Gemini CLI (original report, opens in a new tab)
AIR Security researchers disclosed a flaw letting a plugin repository owner swap already-approved code past SHA-pinning checks with no user click, affecting four major coding agents; Anthropic patched in Claude Code 2.1.179 and OpenAI in Codex 0.146.0, while Microsoft shipped no Copilot fix and Google retired Gemini CLI uncorrected.
AnthropicOpenAIGitHubMicrosoftGoogle
AI-generated summary. Open the headline to read the original report.
Beyond feature support
Quality evidence.
How well a model performs depends on the task, version, and settings.
We have not published a reviewed quality evaluation for these versions yet. Feature support and price do not establish output quality.
Try the same small task with each option. Check whether the result works, passes your tests, and can be maintained.
Published evaluations to exploreSWE-bench (opens in a new tab)
Check the exact version and test setup. Results on one task do not establish quality on another.