Skip to content
Back

AI posts

AI release notes

Official updates from Anthropic, OpenAI, Google, xAI, Mistral, DeepSeek and Meta, newest first. Open any update for a short AI summary and the full notes.

Compiled by the Toolstash team · Updated

  1. Anthropic · Claude Code

    Claude Code 2.1.291

    2 fixes

    Fixed a regression in 2.1.290 where cloud sessions could drop answers to permission prompts. Fixed a regression in 2.1.288 where the last messages of a session could be lost when quitting.

  2. Anthropic · Claude API

    We’ve added capabilities.server_tools to the Models API

    GET /v1/models and GET /v1/models/{model_id} now report whether each model accepts the web search and code execution tools. To check whether a model accepts the code execution tool, read capabilities.server_tools.code_execution. The top-level capabilities.code_execution reports whether code can call your request’s other tools.

  3. Google · Gemini API

    Deprecation announcement: The gemini-3.1-flash-image model is deprecated. Migrate to gemini-nano-banana-2.1

    Deprecation announcement: The gemini-3.1-flash-image model is deprecated (no shutdown date announced). Migrate to gemini-nano-banana-2.1.

  4. Google · New model

    New model: Gemini Nano Banana 2.1

    Gemini Nano Banana 2.1 generally available (GA): Released Gemini Nano Banana 2.1 (gemini-nano-banana-2.1), the latest high-efficiency image generation and conversational editing model. An update to Nano Banana 2 (gemini-3.1-flash-image), it maintains Flash-level speed and cost efficiency while delivering significant improvements in visual quality, prompt adherence, multi-turn character consistency, text rendering, and wide and panoramic aspect ratio generation (1:4, 4:1, 1:8, 8:1) across 1K, 2K, and 4K resolutions.

  5. Mistral AI · New model

    Mistral Large 4 is now available in Public Preview

    For details, see the Mistral Large 4 model card. Mistral Large 4 is an open-weight, general-purpose multimodal model with a Mixture-of-Experts architecture and a 1M context window. Open weights are coming soon. Launch pricing: 50% off for 2 weeks.

  6. Anthropic · Claude Code

    Claude Code 2.1.290

    15 new130 fixes45 other

    Added serverToolUses to the result of a mod's turn.step hook: the tool calls the API ran itself (the advisor), each with its id, name, input, start and end. Added agentId to the tool.check event of plugin hooks, so a hook can tell a subagent's permission check from the main session's.

  7. Anthropic · Claude API

    We’ve added capabilities.thinking.types.disabled to the Models API

    GET /v1/models and GET /v1/models/{model_id} now report whether each model accepts thinking: {type: "disabled"}, which turns thinking off.

  8. OpenAI · Codex CLI

    Codex CLI 0.160.1

    1 fix

    Preserve SYSTEMROOT, TEMP, and TMP when launching remote stdio MCP servers with explicitly configured remote environment variables, allowing Unix hosts to retain the Windows executor's startup environment.

  9. Anthropic · Claude Code

    Claude Code 2.1.289

    1 new23 fixes3 other

    Added agent.spawn for teammates, one agent id across plugin hook events, and idle and waiting states in $.agent.list().

  10. Anthropic · Claude Code

    Claude Code 2.1.288

    7 new64 fixes18 other

    Added $.ui.selection() for mods: returns the text you last selected in fullscreen mode and, when the selection lies within one transcript row, that row. Added a built-in gh api to cloud sessions whose image has no GitHub CLI, and fixed the built-in sending control characters from file names, jq filters or GitHub errors to the terminal.

  11. OpenAI · Codex app

    ChatGPT for iOS

    New Home Screen widgets show recent tasks from your selected computers. Track remaining usage and reset times with Home Screen and Lock Screen widgets. Added a setting to keep the keyboard open after sending a message. Faster task list refresh when returning to the app, with existing task titles preserved. Fixed missing tasks when loading or scrolling through task lists. Smoother home screen, send, and keyboard transitions. Canceling a queued prompt returns recoverable text and attachments to the composer. Improved transcript stability when selecting text and loading older messages. Worktree creation now respects your computer's configured storage location.

  12. Anthropic · Claude Code

    Claude Code 2.1.287

    8 new66 fixes32 other

    Added Claude Mods: plugins may now modify deeper behavior. Added You should know, a built-in mod where a side agent watches your back and flags things you or Claude might miss. Turn it on with /plugin enable cc-plugin-you-should-know@builtin (for first-party sessions with telemetry on).

  13. Anthropic · Claude API

    We’ve added a line field to the Models API

    GET /v1/models and GET /v1/models/{model_id} now return the model line each model belongs to. Claude Opus 4.5 and Claude Opus 4.6 both report opus, for example. Use line to group models without parsing their IDs. line is null for a model that belongs to no line. Dreams (research preview) now supports Claude Opus 5.5, Claude Fable 5.1, and Claude Sonnet 5.5.

  14. OpenAI · Codex app

    Codex app

    Added the option to move selected chats to a project at once. Performance improvements and bug fixes. Improved recovery when attaching managed worktrees to chats. Fixed large diffs appearing blank at some zoom levels. Improved retries after queued messages fail to send or chats fail to open.

  15. OpenAI · Codex CLI

    Codex CLI 0.160.0

    4 new6 fixes3 other

    Browse older tasks in the agent command center with a keyboard-accessible “Show more” action. Select transcript text and paste with middle-click in fullscreen mode on supported local Linux X11 terminals.

  16. xAI · Grok API

    Custom moderation controls

    You can now set content guidelines on text API requests. Add a guidelines array to a Chat Completions or Responses API request to set how Grok handles content in the nsfw, violence, and profanity categories. Grok follows the guidelines as it writes its response.

  17. xAI · Grok API

    Grok Voice 2.0 deprecation

    grok-voice-2.0 is deprecated. To upgrade, set model to grok-voice-3.0 or grok-voice-latest, at the same price.

  18. xAI · New model

    Grok Voice 3.0 is available

    grok-voice-3.0 is now available with Speech to Speech, at the same price as Grok Voice 2.0. grok-voice-latest now points to grok-voice-3.0, so sessions that use the alias are upgraded automatically. The API is unchanged.

  19. xAI · New model

    grok-voice-transcribe-1.0 end of life

    grok-voice-transcribe-1.0 is deprecated and reaches end of life on October 2, 2026. All requests to that slug are routed to grok-voice-transcribe-2.0 at the same price, with higher accuracy.

  20. Anthropic · Claude Code

    Claude Code 2.1.286

    7 new49 fixes32 other

    Added a count such as "2 of 5" to the permission prompt when several permission requests stack up. Added mouse support for the "N more" rows of lists in fullscreen mode: click one to jump to that end of the list, with hover and pressed states.

  21. Anthropic · Claude API

    In Python SDK 1.10.0, TypeScript SDK 0.130.0, C# SDK 12.52.0, Go SDK 1.77.0, Java SDK 2.67.0, PHP SDK…

    These releases also keep client.beta.organization and ant beta:organization. We announced the deprecation of the Claude Sonnet 4.5 model (claude-sonnet-4-5-20250929), with retirement on the Claude API scheduled for November 30, 2026.

  22. OpenAI · Codex CLI

    Codex CLI 0.159.3

    1 new

    Eligible local sessions signed in with ChatGPT can now show optional reminders to complete account security setup.

  23. Google · Gemini app

    Let skills in Gemini tackle your most repetitive tasks

    What: Today we’re rolling out skills directly into the Gemini chat globally 1. Skills are a reusable set of instructions that tell Gemini how to handle specific tasks. Think of them as saved shortcuts to faster prompting. Gemini can now help you build custom skills from your chats and automatically run them whenever you enter a matching prompt. For bigger tasks, you can stack multiple skills together at the same time — like combining a writing style skill with a brand guidelines skill to create on-brand assets in your voice. Skills will soon replace Gems as the tool for tailoring instructions for specific tasks. To make this transition seamless, we will automatically migrate your Gems into skills when Gems go away. You can learn more about how to transition your Gems into skills by visiting our help center. Why: We want to help you spend less time prompting and more time doing. With skills, you can save your most-used instructions once and run them endlessly – so you can skip the repetitive prompting and jump straight to the results.

  24. Google · Gemini CLI

    Gemini CLI 0.62.0

    12 fixes7 other

    add early return on unsupported store in tasks metadata endpoint. format MCP tool call titles as structured signatures and segregate explanations.

  25. Anthropic · Claude Code

    Claude Code 2.1.285

    10 new86 fixes40 other

    Added CLAUDE_CODE_DISABLE_WEB_FETCH environment variable to turn off the WebFetch tool. Added claude --desktop to open the Claude desktop app on the current directory, or on a session with --continue / --resume <id>.

  26. OpenAI · Codex app

    Codex app

    Performance improvements and bug fixes. Improved terminal startup in attached worktrees and fixed failures caused by unavailable worktree environment settings. Kept earlier answers visible when you send new instructions during a task. Fixed automatic scrolling and prevented keyboard navigation through messages from jumping back to the latest response. Improved startup recovery when you return to the app window.

  27. OpenAI · Codex CLI

    Codex CLI 0.159.0

    6 new6 fixes2 other

    Opt-in instant_interrupt lets new input steer Codex during model responses or long-running code-mode calls. New sessions get a compact welcome screen and consistent headers, with occasional tips during and after turns.

  28. OpenAI · Codex CLI

    Codex CLI 0.159.1

    1 new

    Added GPT-6.1 Sol as the default model in the bundled catalog and Amazon Bedrock Mantle and Runtime catalogs.

  29. OpenAI · Codex CLI

    Codex CLI 0.159.2

    1 fix

    Suppressed console windows flashing on Windows when Codex launches background processes and sandboxed commands.

  30. OpenAI · New model

    GPT-6.1 Sol in Codex and ChatGPT Work

    GPT-6.1 Sol offers near-Astra performance for complex work at a lower cost than Astra. Consider it for repeated, long-running work across code, apps, and documents. Availability depends on your plan, client, and workspace settings. Use gpt-6.1-sol.