OpenAI Articles
48 articles across AI Catchup's news, guides, tutorials, and comparisons.
All OpenAI articles
OpenAI Brings GPT-5.6 Model Family to Kiro for Spec-Driven Coding
OpenAI says the GPT-5.6 model family, including Sol, Terra, and Luna, is now available in Kiro. The OpenAI-AWS update targets structured, long-running software work and reports roughly an 82% cost reduction per successful Terminal-Bench 2.1 task for GPT-5.6 Terra in Kiro's spec-driven environment.
ChatGPT Sites Adds Teammate Editors: Collaborate on the Same Project and Publish Together
OpenAI added teammate editing to ChatGPT Sites. A Site owner can give an active member of the same ChatGPT workspace Can edit access; editors can update and save the Site, then publish later versions to the same URL after the owner’s first publish. Ownership, access, URL, secrets, and custom-domain controls remain with the owner.
ChatGPT Voice Can Start and Steer Codex Tasks on Desktop, With Mobile Remote Access
ChatGPT Voice now provides a hands-free way to start, monitor, and steer Codex work. OpenAI says Voice can coordinate Codex tasks in the ChatGPT desktop app, while mobile access is available through Remote on iOS after pairing an iPhone with a desktop host. Voice conversations use the Codex task’s usage budget and require a new chat or task to begin in voice mode.
OpenAI Adds API-Key Spend Dashboards and Hard Limits
OpenAI's API platform now lets teams group Usage and Costs data by API key and set monthly organization or project spend limits. The dashboard feature helps identify which apps and workloads drive spend, while a hard limit can stop affected API traffic with a documented 429 error after the cap is reached.
OpenAI Adds Transparent Backgrounds to GPT-Image-2 in API Preview
OpenAI added transparent-background output in preview for gpt-image-2 and gpt-image-2-2026-04-21 across the Images API and Responses API image-generation tool, with PNG or WebP output required.
Codex Opens GPT-5.6 Sol's 1M-Token Context to ChatGPT Accounts
GPT-5.6 Sol's 1M-token context window in Codex is now available for usage through ChatGPT accounts, not only API keys, according to OpenAI developer Tibo Sottiaux. The announcement also repeats a warning that Codex's default context length is tuned for performance and cost.
OpenAI's GPT-5.6 Builder's Guide Shows How to Make Agents More Efficient
OpenAI's GPT-5.6 builder's guide lays out a Responses API architecture for longer-running agents: persist reasoning, compact context, delegate work across agents, and move deterministic tool processing into code. OpenAI reports that these patterns raised GPT-5.6 Sol's ARC-AGI-3 score from 13.3% to 38.3% while using roughly 6x fewer output tokens.
ChatGPT Adds Interactive Quizzes and Restaurant Reservation Search
OpenAI has added interactive quizzes to ChatGPT on web and mobile for consumer and Edu plans, while restaurant reservation search is rolling out across ChatGPT plans on mobile, web, and desktop. ChatGPT can use your topic, location, date, party size, and preferences to help with each workflow.
OpenAI Launches Computer History for ChatGPT and Codex on Mac
OpenAI is rolling out Computer History in the ChatGPT desktop app for macOS. When users opt in, ChatGPT and Codex can use recent interaction events to build local memories and a timeline, answer questions about past work, and suggest skills or automations. The feature is off by default and includes app and website permissions, pause and delete controls, and a 48-hour temporary event-file window.
Prompting GPT-5.6: Message Roles, Effort, and Agentic Prompts That Work
GPT-5.6 rewards precise, explicit prompts: structure developer messages as identity, instructions, examples, then context, keep stable content first for prompt caching, and pick reasoning effort deliberately (xhigh for complex multi-step work). Move saved prompt objects into code before OpenAI shuts down v1/prompts on November 30, 2026.
ChatGPT Desktop App Arrives on Linux in Preview
OpenAI is previewing the ChatGPT desktop app for Linux. The app brings ChatGPT, ChatGPT Work, and Codex to supported Linux systems, with official posts listing Ubuntu 24.04 and 26.04, Debian 13, and Fedora 43 and 44, plus .deb and .rpm packages for x64 and ARM64.
Codex Can Import Workflows From Other AI Agents
OpenAI now lets the ChatGPT desktop app and Codex CLI import supported setup and recent work from other agents. The desktop app can import from Claude Code, Claude Cowork, and Cursor, while Codex CLI supports Claude Code and Cursor, with automatic updates and import history available in the desktop app.
OpenAI Expands Daybreak With GPT-5.6-Cyber
OpenAI expanded its Daybreak cybersecurity program with Blue and Red access tiers and introduced GPT-5.6-Cyber, a purpose-trained model for authorized vulnerability research, exploit validation, and security testing. Access is limited to approved defenders and organizations.
OpenAI Resets ChatGPT Work And Codex Limits For Paid Users
OpenAI executive Tibo first announced a usage reset for paid ChatGPT Work and Codex users on August 8, 2026, then said the five-hour limit will return for Plus accounts across both products on August 26. The approved posts do not publish a numeric allowance, a new API billing rule, or details beyond that plan and product scope.
OpenAI Brings Codex Security Review to GitHub Pull Requests
Codex Security Review is OpenAI’s research-preview workflow for deeper security analysis of GitHub pull requests. It uses the diff, repository context, and optional threat-model guidance, then reports actionable findings in the pull request and a fuller report in Codex. Enterprise, Business, Edu, and Pro users can configure it; Plus is excluded.
ChatGPT Expands GPT-5.6 Luna Access and Adds a Reasoning Slider
OpenAI is making GPT-5.6 Sol the single Chat experience for Plus and Pro users, adding a reasoning-effort slider across web, mobile, and desktop. Free and Go users are getting GPT-5.6 Luna as the default model, unlimited text chats, and a Think button for harder questions.
Agent Plugins: An Open Standard for Skills and MCP
OpenAI Developers introduced Agent Plugins, an open, vendor-neutral package format for sharing Agent Skills and MCP server configurations across compatible agent clients.
OpenAI Details GPT-Live’s Full-Duplex Voice Architecture
OpenAI says GPT-Live is a third-generation voice system that can listen and speak at the same time, keep media flowing while deeper reasoning and tool use run asynchronously, and power ChatGPT Voice today. The architecture is also intended to underpin an upcoming GPT-Live API.
GPT-5.4 Leaves ChatGPT-Signed-In Codex on August 31
OpenAI says GPT-5.4 and GPT-5.4 mini will no longer be available to users signed in with ChatGPT starting August 31, 2026. The models remain available through the OpenAI API and Codex sessions authenticated with an API key.
OpenAI Releases Codex Security CLI for Repository Scans and CI Checks
OpenAI says its open-source Codex Security CLI can scan repositories, track findings across runs, verify fixes, and add security checks to CI/CD. The beta CLI requires Codex Security access and is built for teams that want code-aware security review in the terminal.
Codex ImageGen Adds a Lightbox + Canvas Workflow for Image Editing on Desktop
OpenAI Developers announced a new ImageGen workflow in Codex: a lightbox viewer and a canvas-style editing surface designed for direct, point-and-edit changes (erase, annotate, place text) instead of long prompt rewrites.
OpenAI Cuts GPT-5.6 Prices and Renames Priority Processing to Fast Mode
OpenAI reduced GPT-5.6 Luna pricing by 80% and Terra by 20%, then announced temporary GPT-5.6 Sol API and credit pricing relief for the next three months. OpenAI also renamed Priority processing to Fast mode and updated its pricing guidance for the GPT-5.6 family.
OpenAI’s GPT‑Live‑1 powers the new ChatGPT Voice experience (web search, memory, and multimodal chat)
OpenAI’s latest ChatGPT Voice rollout is powered by GPT‑Live‑1 for paid users and GPT‑Live‑1 mini for Free users, enabling simultaneous listening and speaking and adding support for web search, memory, and mixed text+image conversations in Voice.
OpenAI Introduces GPT-Red: An Internal Automated Red-Teaming Model for Prompt Injection
OpenAI published details on GPT-Red, an internal automated safety red-teaming model designed to find prompt injection vulnerabilities at scale and strengthen defenses before broader deployment.
Cursor Adds GPT-5.6 Sol, Terra, and Luna
Cursor says the GPT-5.6 model family (Sol, Terra, Luna) is now available in Cursor and selectable from the model picker, with a published CursorBench score for Sol.
OpenAI Launches ChatGPT Work: A Codex-Sibling Agent for Long-Running Deliverables
OpenAI introduced ChatGPT Work, a new agent in ChatGPT powered by Codex and GPT-5.6 that takes action across apps and files and stays on a project for hours to turn a goal into finished work. Work handles research and deliverables; Codex stays the dedicated software-development agent.
OpenAI adds GPT-Realtime-2.1-mini to the API with reasoning and tool use
GPT-Realtime-2.1-mini is a distilled reasoning model for realtime voice, with function calling, a 128,000-token context window, and text, audio, and image input. OpenAI's pricing page lists it at the same rates as GPT-Realtime-mini: $0.60/$2.40 per 1M text tokens and $10/$20 per 1M audio tokens.
OpenAI introduces GeneBench-Pro, a research-level benchmark for agentic computational biology
On June 30, 2026, OpenAI announced GeneBench-Pro: a research-level benchmark meant to measure how well AI agents navigate messy biological data and make the judgment calls real computational biology depends on. OpenAI says GeneBench-Pro contains 129 questions and is open-sourcing 10 representative case studies as a public package on Hugging Face under the MIT License.
Codex Permission Profiles: Least-Privilege Controls for Local Agent Work
OpenAI shipped Codex permission profiles in beta -- reusable, inheritable policies that replace the coarse sandbox_mode/sandbox_workspace_write combo. A profile binds OS-enforced filesystem read/write/deny rules (down to **/*.env) to per-domain network and Unix-socket rules. Enterprise admins get fail-closed allowlists via requirements.toml. Profiles govern local sandboxed command execution only, not MCP servers, app connectors, browser, or cloud.
OpenAI Previews GPT-5.6: Sol, Terra, and Luna in Limited Preview
OpenAI announced a limited preview of the GPT-5.6 family: Sol, a next-generation frontier flagship OpenAI calls a step function better than GPT-5.5; Terra, a balanced model competitive with GPT-5.5 at 2x lower cost; and Luna, its most cost-efficient model. Access starts with trusted partners in Codex and the API.
OpenAI Ships a New GPT-5.5 Instant in ChatGPT: Better Intent, Constraints, and Shopping/Local Recs
OpenAI says a new version of GPT-5.5 Instant is rolling out in ChatGPT with better intent understanding, more reliable handling of complex constraints, and improved shopping and local recommendations. Per OpenAI, it reaches paid users first and free users the next day. GPT-5.5 Instant is ChatGPT's default for logged-in users.
OpenAI Codex Adds Record & Replay: Turn a Demonstrated Mac Workflow Into a Reusable Skill
OpenAI shipped Record & Replay in Codex app 26.616, a macOS feature that turns a workflow you demonstrate into a reusable skill. It builds on Computer Use, which you or your administrator must enable, and is unavailable at launch in the European Economic Area, the United Kingdom, and Switzerland.
OpenAI: Responses API web search can now return image results
OpenAI added image results to the Responses API web_search tool, letting apps retrieve web-grounded visuals (with source links) alongside regular text results.
ChatGPT 'Dreaming': OpenAI's New Memory Architecture Curates What It Remembers in the Background
OpenAI rolled out a more capable, compute-efficient ChatGPT memory architecture built on 'dreaming' -- a background process that curates memories by referencing chat history without prompting. It carries context forward better, follows preferences across conversations, and updates memories as time passes. Plus and Pro users in the US first, with Free and international users following.
Codex for Every Role: Role-Specific Plugins, Codex Sites, and Annotations Beyond Code
OpenAI is pushing Codex past software development with three releases: six role-specific plugins bundling 62 apps and 110 skills, Codex Sites that turn analysis into shareable hosted web apps in preview for business and enterprise, and annotations that now refine documents, spreadsheets, and presentations -- not just code and websites.
Codex Build iOS Apps Plugin: Mirror the Simulator in the Browser and Hot-Reload SwiftUI Previews
OpenAI's Build iOS Apps plugin lets Codex mirror the iOS Simulator in the in-app browser and hot-reload package-backed SwiftUI previews without leaving Codex. It packages Swift and iOS workflows -- designing App Intents and Shortcuts, building and refactoring SwiftUI, auditing performance, and debugging on simulators through XcodeBuildMCP-backed flows. The plugin is open source in OpenAI's plugins repo.
Codex Adds Windows Computer Use + ChatGPT Mobile Windows Connections for On-the-Go Steering
OpenAI says Codex can now use Computer use on Windows to test apps, debug flows, and review work on your Windows machine, and that Codex in the ChatGPT mobile app can connect to Windows machines so you can steer tasks from your phone.
OpenAI's Secure MCP Tunnel: Connect Private MCP Servers Over Outbound-Only HTTPS
OpenAI's Secure MCP Tunnel connects private and on-prem MCP servers to ChatGPT, Codex, and the Responses API without opening inbound firewall ports. You need a tunnel_id, an API key with Tunnels Read plus Use, and -- separately -- ChatGPT developer-mode access. That permission split is the most common setup blocker.
OpenAI Codex adds Locked computer use on Mac (keep Computer Use running after lock)
Codex can now keep using Mac apps after your screen locks. Locked computer use is a narrow, Codex-only unlock path with a short-lived authorization window and safeguards like relocking on local input.
Codex in the ChatGPT Mobile App: Run, Review, and Steer Codex from Your Phone
Yes, you can drive Codex from your phone. The ChatGPT mobile app on iOS and Android connects to a host running the ChatGPT desktop app on macOS or Windows, and from the phone you start work, review diffs and terminal output, and approve actions. Codex still executes on the host, which must stay awake and online.
Codex Chrome Extension: How Codex Drives a Signed-In Browser for LinkedIn, Salesforce, Gmail, and Internal Tools
OpenAI's Codex Chrome extension lets the agent use Chrome for browser tasks that need signed-in state -- LinkedIn, Salesforce, Gmail, internal tools. Available in the Codex app in all regions except EU and UK at launch. Setup is Codex > Plugins > add Chrome > install extension > approve permissions. Invoke with @Chrome. By default Codex asks before each new website; allowlist/blocklist and elevated-risk options live in Computer Use settings.
Codex Hooks and Programmatic Access Tokens: Setup, Trust Model, and What Actually Runs Today
Codex access tokens are ChatGPT Business and Enterprise workspace credentials for non-interactive Codex CLI runs. Create one in the ChatGPT admin console at chatgpt.com/admin/access-tokens, then authenticate with the CODEX_ACCESS_TOKEN environment variable or codex login --with-access-token. Hooks are the in-session extensibility framework: eleven lifecycle events, of which only command handlers execute today.
Codex CLI 0.130.0 Adds `remote-control`, Richer Plugin Sharing Metadata, and Better App-Server Thread Paging
OpenAI shipped Codex CLI 0.130.0 in May 2026. The release adds a new `codex remote-control` command for starting a headless, remotely controllable app-server, improves app-server clients with paging options for large threads (unloaded/summary/full turn items), expands plugin sharing with link metadata and discoverability controls, adds Bedrock auth support for AWS console-login credentials from `aws login` profiles, and fixes several app-server/thread reliability issues.
Codex CLI 0.129.0 Adds Modal Vim Composer, Redesigned Resume/Fork Picker, and a `/hooks` Browser
OpenAI shipped Codex CLI 0.129.0 on May 7, 2026. The release brings modal Vim editing to the TUI composer via `/vim`, a redesigned resume/fork picker with raw scrollback and workspace-aware `/diff`, a new `/hooks` browser with before/after compaction support, expanded plugin management with workspace sharing and access controls, theme-aware status lines, and Codex Apps auth surfaced through Guardian. Plus a long bug-fix list across Linux/Windows sandboxes, MCP, and TUI input handling.
Codex CLI 0.128.0 Lands Persisted `/goal` Workflows: Ralph-Style Agents That Don't Stop Until Done
OpenAI shipped Codex CLI 0.128.0 in April 2026 with a new `/goal` system that keeps a goal alive across turns and runs the agent until it is achieved. The release adds app-server APIs and model tools for goals, runtime continuation, and TUI controls to create, pause, resume, and clear goals. Felipe Coury credits co-worker Eric Traut (the Pyright lead) for the design, and frames it as Codex's take on the Ralph loop pattern.
GPT-5.5 Is Here: State-of-the-Art Agentic Coding, 1M Context, and a New Pro Tier
OpenAI launched GPT-5.5 on April 23, 2026 -- its smartest model yet, with state-of-the-art scores on Terminal-Bench 2.0 (82.7%), GDPval (84.9%), and OSWorld-Verified (78.7%), GPT-5.4 per-token latency, and a new GPT-5.5 Pro tier for harder work. Available in ChatGPT and Codex today, with API at $5/M input and $30/M output coming soon.
OpenAI Codex Goes 'For Almost Everything': Mac Computer Use, Browser Comment Mode, and Thread Automations Explained
OpenAI shipped a major Codex update on April 16, 2026 that pushes the product past coding into general work. Three changes matter: Codex can now drive your Mac apps directly, an in-app browser captures both screenshots and DOM elements through 'comment mode', and Codex threads can run continuously to watch Slack, email, and PRs. Here is how each works, the workflows that justify each one, and where Codex now sits relative to Perplexity Personal Computer and Claude Code Routines.
Prompt Engineering in 2026: The Playbook That Works Across Claude and GPT
Prompt engineering in 2026 has two layers: cross-model fundamentals (clear direct instructions, motivated rules, 3-5 structured examples, XML boundaries, long documents before the question) and model-specific overrides that matter more than ever, because the newest models need instructions deleted as often as added. Start here, then apply the guide for your model.