What changed, on every surface, within minutes.

Feature · API: v1/responses You can now choose 12-hour prompt caching in the API. 12-hour prompt caching keeps shared instructions, tools, and context cached longer to benefit from lower cached-input pricing, even when requests are hours apart. 12-hour cache writes cost 2.5x the input token price. Read more here.

• gpt-3.5-turbo-0125 | gpt-3.5-turbo, gpt-3.5-turbo-completions
• gpt-4-0613 | gpt-4, gpt-4-0613-completions, gpt-4-completions
• gpt-4-turbo | gpt-4-turbo-2024-04-09, gpt-4-turbo-completions
• gpt-4.1-nano | gpt-4.1-nano-2025-04-14

On May 7th, 2026, we notified developers using OpenAI’s self-serve fine-tuning platform of updates to availability. Inference on fine-tuned models will continue to be available until the base models are deprecated.

Feature · Model: gpt-live-1 · API: v1/audio/voices · API: v1/live/sessions Released prompt-to-voice for all API customers. Create a custom voice from a text description through the API or voice Playground, then reuse it in GPT-Live sessions. Creating a prompt-generated voice is free; speech generation uses regular GPT-Live session billing. Added nine…

Feature · API: v1/agents OpenAI-hosted sandboxes in the Agents API now include the Artifact SDK and skills for creating and editing spreadsheets, presentations, and Word documents. Added Ultrafast mode to the Agents API. Set agent.service_tier to ultrafast with a supported model to reduce the time between generated tokens.

• Open /config to search preferences across categories, change supported settings immediately, and see which changes require a restart. (#52990, #53429)
• Send a follow-up while Codex is responding to steer it immediately. Interrupting a turn also stops active Code Mode cells and their nested tool calls. (#51812, #51835)
• Configure a leader key in /keymap. The default is Ctrl+X, shortcut hints stay open until you choose or cancel, and voice mute moves to F9. (#52273)
• Restart a managed background server from /daemon. Codex warns about work the restart could interrupt, then reconnects automatically. (#53173)

Feature · Model: gpt-live-1 · Model: gpt-6-luna · API: v1/live/sessions Added native Decisions integration in GPT-Live. Your app can act while your user is speaking or react to what Live says. Let Live call Decisions directly through native delegation, or trigger decisions as input or output transcript deltas arrive.

• Open /config to search preferences across categories, change supported settings immediately, and see which changes require a restart. (#52990, #53429)
• Send a follow-up while Codex is responding to steer it immediately. Interrupting a turn also stops active Code Mode cells and their nested tool calls. (#51812, #51835)
• Configure a leader key in /keymap. The default is Ctrl+X, shortcut hints stay open until you choose or cancel, and voice mute moves to F9. (#52273)
• Restart a managed background server from /daemon. Codex warns about work the restart could interrupt, then reconnects automatically. (#53173)

On October 14, 2026, we notified developers using certain GPT-4 series models of their deprecation and removal from the API on January 14, 2027. Fine-tuned versions of these models follow the same shutdown schedule as their underlying base models.

Effective October 14, 2026, at noon PDT (19:00 UTC), GPT-5.5 retires from Codex for users signed in with ChatGPT. The OpenAI API and Codex sessions authenticated with an API key aren't affected. Use GPT-6.1 Sol (gpt-6.1-sol) when available to your plan and client. Update workspace defaults, saved model settings, managed configurations, custom agents…

• Fixed a TUI crash when asynchronous questions contain multiple lines, preserving line breaks and complete hyperlink destinations. (#51866)
• Fixed startup failures caused by differences between a running background server's feature settings and CLI defaults. Compatibility checks now apply only to explicit command-line feature overrides. (#52648)

Feature · API: v1/agents You can now pre-warm an OpenAI-hosted sandbox before attaching it to an Agents API session. Prewarming improves latency by creating and configuring the sandbox before a user finishes sending their message. This means the user doesn't have to wait for all this work to complete. OpenAI-hosted sandbox compute is now billed during…

Composer predictions adds autocomplete-style suggestions for your next message in the Codex desktop app. After Codex responds, it uses the current conversation to suggest a follow-up prompt in the message box. Press Tab to accept the suggestion as an editable draft, or type your own message. Accepting a prediction doesn't send it for you. The feature is…

• Fixed a TUI crash when asynchronous questions contain multiple lines, preserving line breaks and complete hyperlink destinations. (#51866)
• Fixed startup failures caused by differences between a running background server's feature settings and CLI defaults. Compatibility checks now apply only to explicit command-line feature overrides. (#52648)

• Add tools for creating and listing managed Git worktrees from trusted local projects when the worktrees feature is enabled. (#50148)
• Pin tasks in the agent Command Center with p and keep them in a shared Pinned group when supported by the server. (#51500)
• Navigate and copy transcript blocks with /copy, use Ctrl+Insert to copy selections, and tune mouse-wheel scrolling with tui.mouse_scroll_speed. (#50434, #50215, #50209)
• Make URLs clickable in approval headers, questions, MCP prompts, warnings, banners, and verification prompts, including when links wrap across lines. (#51439, #51449, #51450, #51451, #51452, #51458)

Feature · Model: gpt-6.1-sol · API: v1/responses Added Ultrafast mode for GPT-6.1 Sol in the Responses API. Use gpt-6.1-sol with service_tier: "ultrafast" to reduce the time between generated output tokens. It is available to all API users, subject to rate limits, with global processing and US and EU data residency. See Ultrafast pricing.

• Fixed queued messages getting stuck on Sending and restored Steer in
• Improved scrolling and navigation between files in the Changes panel.
• Added clearer recovery guidance when a tool call needs approval but the chat's

• Add tools for creating and listing managed Git worktrees from trusted local projects when the worktrees feature is enabled. (#50148)
• Pin tasks in the agent Command Center with p and keep them in a shared Pinned group when supported by the server. (#51500)
• Navigate and copy transcript blocks with /copy, use Ctrl+Insert to copy selections, and tune mouse-wheel scrolling with tui.mouse_scroll_speed. (#50434, #50215, #50209)
• Make URLs clickable in approval headers, questions, MCP prompts, warnings, banners, and verification prompts, including when links wrap across lines. (#51439, #51449, #51450, #51451, #51452, #51458)

Ultrafast mode speeds up token generation with GPT-6.1 Sol in Codex and ChatGPT Work. Ultrafast is available on Pro $500 and eligible Enterprise and Edu plans. Enterprise access is off by default; workspace owners must enable it. GPT-6.1 Sol Ultrafast supports inference residency in the United States and Europe (EEA + Switzerland). See Ultrafast mode for…

Update · Model: chat-latest Updated the chat-latest snapshot, which points to the latest model available in ChatGPT for Plus, Pro, Business and Enterprise users. We recommend leveraging the GPT-6 model family for production API usage, but feel free to use this model to test the latest improvements for chat use cases. The underlying model snapshot will be…

• Fixed Windows sandbox refresh failures after app updates.
• Fixed worktree creation and chat forking for WSL repositories whose paths
• Kept the recent chats section collapsed when switching between chats.
• Fixed Remove from section so it removes a chat from its sidebar section

• GPT-6.1 Sol is now the default model in the bundled and Amazon Bedrock catalogs. (#49318, #49339)
• Amazon Bedrock supports multi-agent V2 and Ultra reasoning on compatible models; Bedrock Mantle also accepts AWS GovCloud regions. (#49345, #49813)
• Sign in to MCP servers from an active terminal session with /mcp login <name>. (#49290)
• Choose your microphone, speaker, and microphone input channels for voice conversations, with preferences saved locally. (#49437, #49836)

• Page previews now appear inline in responses.
• Added support for opening Codex task links directly on iOS.
• Redesigned the approval picker with clearer permission choices.
• Find plugins from a single Plugins entry in the composer's plus menu.

Update Simplified API usage tiers from five to three: Build, Launch, and Grow. Organizations automatically upgrade as total credit purchases reach tier minimums. See usage tiers for monthly usage limits and how to view rate limits for each model.

Feature · Model: gpt-6-luna · API: v1/decisions Released the Decisions API in beta with gpt-6-luna. Turn text and images into typed answers 10x faster than the Responses API.

Feature Added an in-product flow for HIPAA compliance support in API Organization settings > General. Admins of eligible organizations can now accept the standard Business Associate Agreement (BAA) and enable HIPAA compliance support for their organization. See the Help Center for eligibility, covered services, and configuration requirements.