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. Google · New model

    New models: Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS

    Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS generally available (GA): Released our next-generation text-to-speech (TTS) audio models and the Gemini API Voices endpoint (/v1beta/voices): Gemini 3.8 Flash TTS (gemini-3.8-flash-tts): Flagship creative TTS model engineered for studio-grade voice fidelity, nuanced acting, regional dialects, and long-form multi-turn stability. Gemini 3.8 Flash-Lite TTS (gemini-3.8-flash-lite-tts): Fast, cost-efficient TTS model built to replace gemini-3.1-flash-tts-preview for high-throughput production and real-time voice agent cascades. Voice design, Voice replication, and the Extended Voice Library: Create persistent custom vocal personas from text prompts, replicate voices with consent verification, and query 150+ prebuilt and custom voices.

  2. Google · Antigravity

    Local model support with LiteRT and LocalOpenAI, standardized evaluation presets, and custom subagent models

    This release announces official Antigravity SDK local model support with first-class LiteRTAgentConfig and LocalOpenAIAgentConfig configurations, introduces a standardized evaluation preset (AgentConfig.eval()) for core coding evaluations, enables custom subagent model overrides, turns on the schedule tool by default across standard tool groups, and pairs background task management with command execution.

  3. Anthropic · Claude Code

    Claude Code 2.1.278

    1 new1 other

    Added an Auto mode server row to /status showing whether this session's auto mode classifier runs on the server.

  4. Google · Antigravity

    Inline Kitty graphics for math and Mermaid diagrams, interactive ask_question navigation, dedicated rule…

    Introduces inline Kitty graphics rendering for LaTeX math and Mermaid diagrams in the artifact detail viewer with ASCII fallback and horizontal panning, interactive question navigation with write-in previews in ask_question, dedicated 20,000-token rule budgets to prevent customization eviction, and caps model retry backoff at 30 seconds, alongside plugin lifecycle hardening, Bubble Tea v2.0.9 upgrades, and headless background task logging.

  5. Google · Antigravity

    Windows sandbox support

    This release introduces file and network sandboxing on Windows.

  6. Anthropic · Claude Code

    Claude Code 2.1.277

    10 new57 fixes20 other

    Added AGENTS.md support: in a project with no CLAUDE.md, Claude Code reads AGENTS.md instead; change it under "Project instructions" in /config (not yet on Bedrock, Vertex or Foundry). Added CLAUDE_GATEWAY_PROXY_IS_EGRESS_BOUNDARY=1 for Claude apps gateways whose only egress is a forward proxy: every outbound request hands the proxy the hostname instead of resolving it locally.

  7. Anthropic · Claude Code

    Claude Code 2.1.276

    1 fix

    Fixed every request failing with 400 … Input tag 'advisor_20260301' when ANTHROPIC_BASE_URL points at a proxy or gateway (2.1.275 regression).

  8. Anthropic · Claude API

    For cache diagnostics, a response to a request that sends the cache-diagnosis-2026-04-07 beta header now…

    The field is null when the request did not include the diagnostics object. Previously the field was omitted in that case. The Compliance API local session endpoints now also return transcripts of Claude in Chrome sessions (product_surface value claude_in_chrome), in beta for Claude Enterprise organizations, with your existing Compliance Access Key and the read:compliance_user_data scope.

  9. OpenAI · Codex app

    ChatGPT for iOS

    Create folders directly from the file picker. Added support for writing blocks, with draft alternatives and copy actions. New tasks remember your worktree mode and setup environment for each project. Long queued prompts show a compact preview and expand on tap. Fixed queued prompts missing or appearing outdated after reopening a task. Worktree task cards now open correctly after setup completes. Restored the new-task button for projects containing only worktree tasks. Improved voice-call reliability during temporary connection interruptions. Better alignment of project, environment, and Git controls on iPad. Fixed the thinking timer resetting when a new task starts.

  10. OpenAI · Codex CLI

    Codex CLI 0.155.1

    1 fix

    New local TUI sessions now leave reasoning summaries disabled by default, fixing request rejection by providers that do not support them. Explicit reasoning-summary settings remain respected.

  11. Google · Gemini API

    Gemini 2.5 models access update: To ensure reliable performance for everyone, we are limiting access to the…

    Gemini 2.5 models access update: To ensure reliable performance for everyone, we are limiting access to the 2.5 models to users who have actively used them in the past. These models are not deprecated and will continue to be served until further notice through the API. For any new projects, use our latest models: 3.5 Flash-Lite or 3.8 Flash. This helps us maintain sufficient capacity for both ongoing legacy workflows and new applications.

  12. Google · Antigravity

    Session-scoped Remote Control, unlimited headless run timeouts, structured error reporting, and artifact…

    Introduces session-scoped Remote Control via --remote-control and /remote-control, lifts the default 5-minute timeout on headless runs, adds structured AGY_ERROR output on stderr, improves word-wise navigation in the prompt editor, and fixes Remote Control workspace permission inheritance, oversized file edit storage limits, full-screen artifact text selection, queued /model switching, and /rewind snapshot resolution.

  13. Google · Antigravity

    Custom agent controls and keyboard navigation

    Custom agents can now switch off the default prompts and tools, and the agent you pick sticks with you after a reload. This release also adds Page Up, Page Down, Home, End, and Escape for moving around a conversation, speeds up switching conversations in the sidebar, and fixes a batch of settings and conversation state issues.

  14. Anthropic · Claude Code

    Claude Code 2.1.275

    13 new61 fixes22 other

    Added the signed-in account to Claude apps gateway sign-in: when the gateway names it, you confirm it before the credential is saved, and /status shows it. Added a send-now key (ctrl+enter, or ctrl+x ctrl+s) that interrupts the current turn and sends all queued messages at once; sent and queued messages show in gray until the model receives them.

  15. Anthropic · Claude Code

    Claude Code 2.1.274

    13 new71 fixes24 other

    Added a visible warning when memory usage is critical, with steps to free memory or restart safely. Added CLAUDE_CODE_MCP_STARTUP_WAIT_MS to bound how long the first non-interactive turn waits for connecting MCP servers (0 = don't wait).

  16. OpenAI · Codex app

    Codex app

    Performance improvements and bug fixes. Fixed WSL terminals opening in the wrong Linux distribution or project folder. Fixed adding WSL project folders, including folders on mapped drives. Updated sign-in screens to show only the methods allowed by your organization's policy.

  17. OpenAI · Codex CLI

    Codex CLI 0.155.0

    6 new6 fixes1 other

    Added experimental /voice conversations with live transcripts and microphone controls on supported builds, enabled through /experimental. The TUI now shows live reasoning summaries in the status row and completion timestamps after successful turns.

  18. Google · Gemini API

    Antigravity Agent 09-2026: Released antigravity-preview-09-2026, which replaces and deprecates…

    Antigravity Agent 09-2026: Released antigravity-preview-09-2026, which replaces and deprecates antigravity-preview-05-2026. If you run on a remote sandbox (environment: "remote") and read only output_text or model_output steps, update the agent string and nothing else changes. If you run tools locally (local_environment) or parse function_call steps, the built-in tools changed. Parameters use PascalCase instead of snake_case, and file edits use line-range replacements instead of full rewrites. Capability 05-2026 09-2026 File creation write_file(path, content) write_to_file(TargetFile, CodeContent, Overwrite, Description) File editing write_file(path, content), full rewrite replace_file_content(TargetFile, StartLine, EndLine, TargetContent, ReplacementContent) File reading read_file(path, offset, limit), byte offsets view_file(AbsolutePath, StartLine, EndLine, ContentOffset) Directory listing list_files(path) list_dir(DirectoryPath) File and code search None, agents used shell commands find_by_name(SearchDirectory, Pattern, MaxDepth) and grep_search(SearchPath, Query, IsRegex) Shell execution code_execution(command, timeout_seconds) Unchanged Web search google_search(queries) Unchanged See the Antigravity Agent guide. antigravity-preview-05-2026 shuts down on October 5, 2026, tracked on the deprecations page.

  19. Google · Antigravity

    Subagent roster prompt injection, background task names, credential invalidation handling, and thinking…

    Introduces automatic subagent roster and instruction injection for custom Markdown agents, preserves explicit background task names across notifications and lists, cleanly clears invalidated auth credentials, and fixes interrupted command exit codes, remote cancellation indicators, artifact viewer window resizing, and frozen thinking frames.

  20. Anthropic · Claude apps

    Claude Cowork comes to every conversation, plus 1 more change

    We're rolling out a new Claude experience that makes everything Claude Cowork does available from any conversation. Ask a quick question or hand Claude a bigger task, like a report, spreadsheet, or presentation, without choosing a mode first. Your chats, Cowork tasks, projects, connectors, and skills carry over. The new experience is rolling out gradually to Pro and Max plans on web, desktop, and mobile. Learn more in Claude Cowork and chat are one Claude. You can now ask for a design, deck, or document in any conversation with Claude, including in Claude Code and the Artifacts tab, and edit it with Claude or directly as you work. Claude Design works inside your conversations with all of its features, including on-canvas editing and importing your design system. Claude Slides gives presentations their own starting point, and Claude Docs is a new way to write living documents with Claude and your team. Artifacts, including Claude Design, Claude Slides, and Claude Docs, are available on every plan, including Free. On Enterprise plans, Claude Design, Claude Slides, and Claude Docs are in beta and off by default until an owner turns them on. Learn more in What are artifacts and how do I use them? and Get started with Claude Docs.

  21. OpenAI · Codex app

    Codex app

    Performance improvements and bug fixes. Fixed back navigation getting stuck between settings and a chat. Kept configuration warnings visible after app startup and window reloads. Improved focus and alignment in answer fields for questions from Codex.

  22. Google · Antigravity

    Live skill reloading, interactive model search, mid-session thought signature fixes, and self-correcting…

    Adds /skills reload for live skill discovery, introduces fuzzy search and autocomplete for /model, improves OAuth scope error guidance, and resolves mid-conversation model switching errors with Gemini thinking models, search_web summarization failures, tool schema validation recovery, and daemon conversation resumption.

  23. Google · Gemini CLI

    Gemini CLI 0.60.0

    1 new10 fixes4 other

    harden path resolution and boundary validation in extension loader.

  24. Anthropic · Claude Code

    Claude Code 2.1.273

    4 new35 fixes25 other

    Added x-claude-code-request-class, x-claude-code-agent-type, x-claude-code-prev-tool-durations, x-claude-code-compaction and x-claude-code-context-compacted request headers for LLM gateways; opt in with CLAUDE_CODE_GATEWAY_HINT_HEADERS=1. Added a notification when an MCP server disconnects mid-session and automatic reconnection gives up, pointing at /mcp.

  25. Anthropic · Claude Code

    Claude Code 2.1.272

    1 other

    Bug fixes and reliability improvements.

  26. Anthropic · Claude appsAnnouncement

    Launching Salesforce in Claude (beta)

    We've launched Salesforce in Claude, a plugin that brings a seller’s accounts, opportunities, and pipeline into Claude with 37 pre-built sales skills. Prep a call, review a deal, create a pipeline dashboard, and send your forecast all from Claude. It’s now available in beta on all paid plans for organizations Salesforce approves through its beta sign-up.

  27. Google · Antigravity

    Copy support for side questions, native backend tool-choice constraints, and subagent MCP inheritance

    Introduces /copy btw to copy side-question responses directly to the clipboard, translates tool-choice constraints to native backend function calling, and fixes plugin hook discovery, subagent MCP server inheritance, relative config path resolution, and premature proactive feedback prompts.

  28. Google · Antigravity

    New Permissions System & Terminal and Git version control for Enterprise

    Enterprise and Business accounts can now use the integrated terminal and Git version control in the sidebar. This release also cuts memory use in long conversations and fixes a range of stability and accessibility issues.

  29. Google · New model

    New models: Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking

    Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking generally available (GA): Released two new audio-to-audio models for real-time voice applications using the Live API: Gemini 3.8 Live (gemini-3.8-live): The default option for most low-latency voice agent experiences and real-time dialogue without reasoning delays. Features interleaved reasoning, default asynchronous function calling, and full session client content updates. Gemini 3.8 Live Extended Thinking (gemini-3.8-live-extended-thinking): High-reasoning audio-to-audio model supporting background reasoning during live audio interactions, recommended when higher background reasoning is required. To get started, see the Live API guide, the Capabilities guide, and the Thinking guide.

  30. Anthropic · Claude Code

    Claude Code 2.1.271

    10 new59 fixes27 other

    Added fast mode in Claude Code Remote sessions (cloud and self-hosted runners): the host's fast-mode setting or /fast typed in the session applies where your organization allows it. Added mouse support to the /config panel in fullscreen mode: the wheel scrolls the settings list, a click on a setting's value changes it, and the row under the pointer is highlighted.