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 2.1.261

    9 new47 fixes11 other

    Added an "Organization policy" line to /status and claude doctor that says why your organization's policy could not be loaded, such as a proxy not passing the endpoint through. Added bashOutputMaxChars and taskOutputMaxChars settings to raise how much command and background-task output Claude receives inline before it is saved to a file, up to 128K characters.

  2. OpenAI · Codex CLI

    Codex CLI 0.153.3

    1 new1 fix

    Added GPT-6-Astra to the Amazon Bedrock model picker for Mantle and Runtime global/US routes.

  3. OpenAI · Codex CLI

    Codex CLI 0.153.4

    2 fixes

    Fixed Astra’s visibility in the bundled model picker and made it the bundled default when no model is explicitly configured. Updated Astra’s guidance to use asynchronous questions only when the tool is available in the session.

  4. Google · Antigravity

    Artifact viewer half-page scrolling, workspace picker grouping setting, and SQLite WAL flushing

    Adds half-page scrolling (Ctrl+D / Ctrl+U) to the artifact viewer, adds the pickerGrouping configuration setting for /resume, improves Mermaid flowchart ASCII rendering, updates unselected model reasoning defaults to medium, and resolves issues with worktree cleanup, SQLite WAL checkpointing on exit, and subagent approval prompts.

  5. Anthropic · Claude Code

    Claude Code 2.1.260

    8 new37 fixes21 other

    Added a diff panel that opens beside the conversation in fullscreen mode and shows your uncommitted changes as Claude edits; toggle it with /diff. Added a likely cause for prompt-cache misses (e.g. tool definitions or system prompt changed, idle past the TTL) to /cost and the status line's prompt_cache field.

  6. Anthropic · Claude API

    Version 1.30.0 of the ant CLI adds ant apply, which creates and updates agents, environments, skills, memory…

    Describe each resource in a file, run ant apply, and approve the plan it prints. Commit the claude-lock.json lockfile it writes so that later runs, on your machine or in CI, update the same resources instead of creating new ones. Per-message effort changes, in beta, are also available on Google Cloud for Claude Fable 5.1, Claude Mythos 5.1, and Claude Opus 5, with the same mid-conversation-output-config-2026-07-01 beta header.

  7. OpenAI · Codex CLI

    Codex CLI 0.153.0

    5 new5 fixes3 other

    Vim mode now supports undo with u and redo with Ctrl+R, preserving complete drafts including pasted content and attachments. The plugin CLI can list, install, and remove plugins from remote marketplaces.

  8. OpenAI · Codex CLI

    Codex CLI 0.153.1

    1 new

    Added support for configuring GPT-6-Astra through the API without changing the default model or showing it in the model picker.

  9. OpenAI · Codex CLI

    Codex CLI 0.153.2

    1 fix

    Corrected the GPT-6-Astra Fast tier description to say “2x speed, increased usage” instead of “1.5x.” This changes only the displayed text, not how requests run.

  10. Google · Gemini API

    Lyria 3.5 generally available: Released the next generation of Google's music generation model: lyria-3.5:…

    Lyria 3.5 generally available (GA): Released the next generation of Google's music generation model: lyria-3.5: Full-length song generation with improved musical coherence, natural vocals, and fine-grained duration and structural control. The model supports text and image inputs and generates high-fidelity 44.1 kHz stereo audio.

  11. Google · Antigravity

    Workspace-grouped resume view, Gemini 3.8 Flash support, and Markdown agent ambient customization inheritance

    Adds an opt-in workspace-grouped view to the /resume picker with Ctrl+F toggling, adds Gemini 3.8 Flash to the model catalog for GEMINI_API_KEY users, updates Markdown-defined custom agents to inherit ambient skills, rules, and subagents by default, and delivers fixes for long MCP OAuth authorization codes, Windows skill path separators, /boost worker tools, and runtime nil pointer crashes.

  12. Google · Antigravity

    Gemini 3.8 Flash using ADC in AGY Enterprise

    Enterprise users can now access Google's flagship Gemini 3.8 Flash reasoning models using ADC.

  13. Anthropic · Claude Code

    Claude Code 2.1.259

    5 new26 fixes6 other

    Added managedMcpServers managed setting: organizations can provide HTTP/SSE MCP servers to every user (same entry shape as .mcp.json); entries that name a command to run are skipped. Added --permission-prompts none for unattended headless hosts: anything that would prompt is denied automatically while the active permission mode (including auto mode) keeps deciding.

  14. OpenAI · Codex app

    Codex app

    Includes updates through September 5, 2026. Performance improvements and bug fixes. Fixed the model picker closing while settings refresh. Distinguished successfully submitted feedback from failed diagnostic uploads. Fixed remote chats getting stuck when reopened after their host changes. Preserved preferences for automatically connecting to SSH hosts after an app restart.

  15. Google · Antigravity

    MCP panel navigation improvements, multi-line comment support in mcp_config.json, and headless stream cleanup

    Improves /mcp panel arrow key navigation and action cycling, adds support for comments and trailing commas in mcp_config.json, ensures headless CLI streams close cleanly on exit, and delivers fixes for duplicate agent listings, deleted working directory startup, and /btw side questions with active goals.

  16. Google · Antigravity

    Quoting, /boost, and improved Settings

    You can now highlight part of a response and quote it as context for a follow-up prompt, and paid users get the new /boost command for a deeper multi-agent reasoning pass. This release also brings UX improvements across settings, the chat panel, and sidebar navigation.

  17. Google · New model

    New model: Gemini 3.8 Flash

    Gemini 3.8 Flash generally available (GA): Released gemini-3.8-flash, our most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows. To get started, see the Gemini 3.8 Flash model page and the Gemini 3.8 Flash guide.

  18. Anthropic · Claude Code

    Claude Code 2.1.258

    2 fixes

    Fixed Claude Code failing to launch on macOS 12 (Monterey), a regression introduced in 2.1.255. Fixed remote and scheduled sessions failing with "user messages must have non-empty content" after a re-sent permission approval could not be applied.

  19. Google · Gemini CLI

    Gemini CLI 0.58.0

    4 fixes3 other

    ensure consistent symlink evaluation in ignore path handling. isolate Docker and container runtime sockets and binaries in macOS Seatbelt.

  20. Anthropic · Claude Code

    Claude Code 2.1.257

    12 new67 fixes25 other

    Added Claude Fable 5.1 (claude-fable-5-1), now the default Fable model — 1M context, $10/$50 per Mtok with $0.25/Mtok cache reads. Added "Time format" (timeFormat) and timeZone settings: 12-hour, 24-hour, 24-hour UTC, or a strftime pattern for the turn-end clock and transcript-view timestamps.

  21. Anthropic · Claude API

    Prompt cache reads on Claude Fable 5.1 and Claude Mythos 5.1 cost $0.25 USD per million tokens: 0.025x the…

    Cache writes are unchanged. On Claude Fable 5.1 and Claude Mythos 5.1, tool_choice types any and tool aren’t supported and return a 400 error. auto and none are unchanged. Thinking blocks produced by Claude Fable 5.1 and Claude Mythos 5.1 are preserved only for the model that produced them or a newer one: earlier models can’t read them, and the API drops one replayed to an earlier model. Per-message effort changes are in beta on Claude Fable 5.1, Claude Mythos 5.1, and Claude Opus 5 on the Claude API. Turn-scoped system messages are in beta (mid-conversation-system-clear-at-2026-08-21 header). thinking.display accepts a third value, "updates", in beta (thinking-display-updates-2026-08-18 header). Text generated by Claude Fable 5.1 and Claude Mythos 5.1 carries Anthropic’s text watermark, and supported image, video, and audio files that Claude produces through the code execution tool carry C2PA Content Credentials when you retrieve them through the Files API on the Claude API. Like Claude Fable 5, both models require 30-day data retention and aren’t available under zero data retention unless expressly authorized by Anthropic. The guides for the Claude Enterprise endpoints of the Admin API (user management and spend limits), the Claude Enterprise Analytics API, and the Compliance API now show the anthropic-version header; send it on every request to these endpoints, as in the rest of the Claude API.

  22. Anthropic · Claude appsAnnouncement

    Claude Fable 5.1 and Claude Mythos 5.1 launch

    We just launched Claude Fable 5.1 and Claude Mythos 5.1, the world’s most advanced models for coding and knowledge work.

  23. Anthropic · New modelAnnouncement

    New models: Claude Fable 5.1 and Claude Mythos 5.1

    We’ve launched Claude Fable 5.1 (claude-fable-5-1), the successor to Claude Fable 5 for long-running agentic coding, knowledge work, and research, alongside Claude Mythos 5.1 (claude-mythos-5-1) for Project Glasswing participants. Both models support a 1M token context window by default, 128k max output tokens, and always-on adaptive thinking, at $10 / $50 USD per MTok, the same as Claude Fable 5, with cache reads cut to $0.25 USD per MTok. Claude Fable 5.1 is available on the Claude API, Claude in Amazon Bedrock, Claude Platform on AWS, Claude on Google Cloud, and Claude in Microsoft Foundry.

  24. OpenAI · Codex app

    ChatGPT for iOS

    Attachments now work across all connected hosts, including Windows and Linux, and support videos from the Photo Library. Press and hold the attachment button to attach recent photos. A new Priority view brings running tasks, unread updates, and tasks awaiting your response to the top of the task list. Queued prompts now sync with the connected host, remain editable, and send even when the app is in the background. Long-running tasks now show their live working time. Task menus now include an option to copy the thread ID. Task list loading and organization are faster and more reliable, with simpler date sections and fewer stalls or disappearing projects. Reconnects are more reliable, resolving stuck Send states, missing approvals, and stale task updates. Long responses now stream with fewer visual interruptions. Side chat messages remain available until you close them, even after the chat can no longer reconnect.

  25. OpenAI · Codex CLI

    Codex CLI 0.152.0

    6 new8 fixes2 other

    Vim mode supports / and ? searches within drafts, highlighted matches, and repeat navigation with n and N. Rate-limit banners offer actions for checking usage, managing credits, resetting limits, and managing plans.

  26. OpenAI · Codex CLI

    Codex CLI 0.152.1

    1 fix

    Guardian approval review now honors Node REPL policies provided through model metadata.

  27. Google · Gemini API

    Agentic video understanding: Released agentic video understanding for Gemini 3.7 Flash, 3.6 Flash, and 3.5…

    Agentic video understanding: Released agentic video understanding for Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite across the Interactions and GenerateContent APIs. The model dynamically navigates video timelines, requesting transcripts, frames, or audio tracks on demand. This approach uses up to 88% fewer tokens for long-form content compared to static processing. To get started, see the Agentic video understanding guide.

  28. Google · Antigravity

    Model autocompletion improvements, subagent streaming optimizations, and MCP tool resolution fixes

    Improves /model autocompletion with Tab acceptance, reduces subagent streaming telemetry overhead, and delivers stability fixes for prompt hooks, MCP tools in subagents, tool call IDs in request history, and OAuth token refreshes.

  29. xAI · Grok API

    grok-imagine-image-quality retirement on November 2

    On November 2, 2026, grok-imagine-image-quality is retired. Requests to the slug will be served by grok-imagine-image-2.0 with quality set to low, with no change to the request or response shape and at a lower per-image price. grok-imagine-image (1.0) is not affected.

  30. xAI · Grok API

    safety\_identifier request field

    You can now send safety_identifier, an opaque end-user identifier assigned by your application, on Chat Completions, the Responses API, deferred chat completions, the Batch API, and the gRPC GetCompletionsRequest. It lets SpaceXSI attribute a policy violation to one of your end users rather than to your API key. The legacy user field is still accepted.