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. OpenAI · Codex CLI

    Codex CLI 0.142.5

    1 fix

    Prevented full Responses WebSocket request payloads from being written to trace logs.

  2. Google · Antigravity

    Subagent & task status bar, editor warnings, and Windows fixes

    New interactive status bar for subagents and background tasks, clipboard image pasting improvements, and Windows fixes.

  3. xAI · Grok API

    Adjustable VAD threshold for Speech to Text

    Speech to Text now accepts a vad_threshold parameter (streaming query param and batch multipart field) to tune the voice-activity gate that skips non-speech audio. Lower values transcribe quieter or noisier speech — useful for narrowband telephony — and 0 disables the gate.

  4. xAI · New modelAnnouncement

    Grok 4.5

    Grok 4.5, SpaceXSI's model for coding, agentic tasks, and knowledge work, is now available on the xAI API. Priced at $2 / 1M input tokens and $6 / 1M output tokens, with configurable reasoning effort (low, medium, or high; default high).

  5. xAI · New model

    Grok 4.5 available in the EU

    Grok 4.5 is now available in the API console for EU users.

  6. xAI · New model

    grok-imagine-video-1.5 modalities

    grok-imagine-video-1.5 now supports text-to-video, image-to-video, and reference-to-video (including optional preset voices), with native 1080p for T2V and I2V. Text-to-video on this model runs as text-to-image then image-to-video under the hood.

  7. xAI · New modelAnnouncement

    Grok Voice 2.0 is available

    grok-voice-2.0 is now available with Speech to Speech. grok-voice-latest will route to this model starting August 5, 2026. To get started, see the Speech to Speech docs.

  8. xAI · New modelAnnouncement

    Grok Voice Think Fast 2.0 is available

    grok-voice-think-fast-2.0 is now available with Speech to Speech. grok-voice-latest will route to this model starting August 5, 2026. To get started, see the Speech to Speech docs.

  9. Anthropic · Claude Code

    Claude Code 2.1.197

    1 other

    Introducing Claude Sonnet 5: now the default model in Claude Code, with a native 1M-token context window and promotional pricing of $2/$10 per Mtok through August 31. Update to version 2.1.197 for access. https://www.anthropic.com/news/claude-sonnet-5.

  10. Anthropic · Claude API

    Claude Managed Agents session event streams now support event deltas

    Opt in with the event_deltas[] query parameter on GET /v1/sessions/{session_id}/events/stream. The event_start and event_delta events preview an agent message’s text as it’s generated, before the complete agent.message event arrives. Listing sessions for Claude Managed Agents now supports backward pagination. When creating a Claude Managed Agents session, you can now override the agent’s configuration for that session. Claude Managed Agents vaults now support an injection_location setting on environment variable credentials (the Environment variable tab). Webhooks for Claude Managed Agents now cover the agent, deployment, and deployment run lifecycle.

  11. Anthropic · Claude appsAnnouncement

    Claude Sonnet 5 launch

    We launched Claude Sonnet 5, our most agentic Sonnet model yet, with substantial improvements over Sonnet 4.6 in reasoning, tool use, coding, and knowledge work.

  12. Anthropic · New modelAnnouncement

    New model: Claude Sonnet 5

    We’ve launched Claude Sonnet 5 (claude-sonnet-5), the next generation of our Sonnet model family, at introductory pricing of $2 / $10 USD per MTok (made the standard price on August 10, 2026). Claude Sonnet 5 supports a 1M token context window, 128k max output tokens, and the same set of tools and platform features as Claude Sonnet 4.6, except Priority Tier, which is not available on Claude Sonnet 5. Three behavior changes apply when migrating: adaptive thinking is now on by default; manual extended thinking (thinking: {type: "enabled", budget_tokens: N}) is removed and returns a 400 error (it was deprecated on Sonnet 4.6); and setting sampling parameters (temperature, top_p, top_k) to non-default values returns a 400 error. Claude Sonnet 5 also uses a new tokenizer that produces approximately 30% more tokens for the same text. The exact increase depends on the content and workload shape. For behavioral differences and model-specific prompting patterns, see Prompting Claude Sonnet 5.

  13. Google · Gemini API

    Released gemini-3.1-flash-lite-image to general availability, our built-in multimodal model optimized for…

    Released gemini-3.1-flash-lite-image (Nano Banana 2 Lite) to general availability (GA), our built-in multimodal model optimized for ultra-low latency and cost-effective image generation and editing.

  14. Google · Gemini app

    Hand off the heavy lifting to Gemini Spark, now available in the macOS app

    What: Gemini Spark, your personal AI agent, is coming to the Gemini app for macOS. It helps you navigate your digital life by working on tasks on your behalf. Operating autonomously and under your direction, it organizes folders, uses your local files to build documents, and handles complex workflows across Google Workspace. Coming soon, you will be able to command Gemini Spark to run tasks or access local files on your Mac directly from the web or mobile app, ensuring your desktop workflows remain at your fingertips wherever you go. Starting today, Gemini Spark is available in English in the Gemini macOS app to Google AI Ultra subscribers aged 18 and over in supported countries. Download the app at gemini.google/desktop. Why: We built this to help you automate your daily busywork, from managing local files to streamlining multi-step projects. With Gemini Spark, we are transforming Gemini from an AI assistant that answers your questions into one that actually works proactively on your behalf.

  15. Google · Antigravity

    Indefinite goals, subagent auto-approvals, and plugin fixes

    Indefinite goals support, subagent auto-approval mode for artifacts, and directory preservation during plugin imports.

  16. Google · New model

    New model: Gemini Omni Flash

    Gemini Omni Flash in public preview: Released gemini-omni-flash-preview, a high-performance multimodal model designed for high-speed video generation and conversational video editing. Using the Interactions API, you can generate 3–10 second videos at 720p from text descriptions or animate still images, and then conversationally edit and refine the outputs. To get started, see the Gemini Omni Flash guide and the Gemini Omni Flash model card.

  17. Mistral AI · New model

    We released Leanstral 1.5, an updated Lean 4 formal proof engineering model with improved SFT mixture…

    This model will be retired on September 30, 2026.

  18. Anthropic · Claude Code

    Claude Code 2.1.196

    3 new16 fixes8 other

    Added support for organization default models — admins set it in the org console; it shows as "Org default" (or "Role default") in /model when you haven't picked one yourself. Added readable default names for sessions at start, making them easier to identify and message.

  19. Anthropic · Claude API

    We’ve removed fast mode for Claude Opus 4.6

    Requests to claude-opus-4-6 with speed: "fast" no longer run at fast speed or premium pricing: they run at standard speed, are billed at standard rates, and do not return an error. The response’s usage.speed field reports the speed used. To continue using fast mode, migrate to Claude Opus 4.8. Read more in Fast mode model availability.

  20. OpenAI · Codex CLI

    Codex CLI 0.142.4

    1 other

    No user-facing changes were identified for this release.

  21. Google · Antigravity

    Optimistic slash commands, strict permission matching, and multiline undo

    Optimistic slash command rendering, strict non-regex permission matching, and unified undo/redo prompt history.

  22. Anthropic · Claude Code

    Claude Code 2.1.195

    1 new8 fixes3 other

    Added CLAUDE_CODE_DISABLE_MOUSE_CLICKS to disable mouse click/drag/hover in fullscreen mode while keeping wheel scroll.

  23. Anthropic · Claude API

    We’ve raised rate limits across the Claude API

    Claude Sonnet and Claude Haiku rate limits now match Claude Opus at every usage tier, and usage tiers have been consolidated into three: Start, Build, and Scale. Most organizations move to a higher tier, no organization receives lower limits than before, and no action is required. You can view your tier and current limits in the Claude Console.

  24. OpenAI · Codex CLI

    Codex CLI 0.142.3

    1 other

    Maintenance-only patch release with no user-facing changes since 0.142.2.

  25. Anthropic · Claude Code

    Claude Code 2.1.193

    6 new5 fixes4 other

    Added autoMode.classifyAllShell setting to route all Bash/PowerShell commands through the auto-mode classifier instead of only arbitrary-code-execution patterns. Added auto-mode denial reasons to the transcript, the denial toast, and /permissions recent denials.

  26. Google · Gemini CLI

    Gemini CLI 0.49.0

    2 new10 fixes15 other

    Support GDC air-gapped Service Identity after auth library update. add eval:inventory CLI command and reporting logic.

  27. Anthropic · Claude API

    We’ve deprecated fast mode for Claude Opus 4.7, with removal on July 24, 2026

    After removal, requests to claude-opus-4-7 with speed: "fast" will return an error. Migrate to fast mode for Claude Opus 4.8. Read more in Fast mode model availability.

  28. Anthropic · Claude apps

    Enable Trusted Devices for Remote Control

    Admins for Team and Enterprise plans can now require members to verify their device before viewing or steering local Claude Code sessions remotely.

  29. OpenAI · Codex app

    Codex Remote reaches general availability

    Codex Remote has reached general availability. Use Codex from the ChatGPT mobile app to start or continue work on a connected Mac or Windows host, review progress, and approve actions from your phone. Remote Control now uses authenticated one-to-one QR pairing between each iOS or Android device and each host. Update the ChatGPT mobile app and Codex App to the latest versions before connecting. Connections used since June 8, 2026, remain paired; older inactive connections need to pair again. The new DigitalOcean plugin lets Codex provision a DigitalOcean Droplet, configure SSH access, and connect it to the Codex App as a remote workspace. Performance improvements and bug fixes. Improved recovery after a remote connection drops and reconnects. Restored the message composer immediately after connecting an app. Reduced browser flashing when Codex takes screenshots.

  30. OpenAI · Codex CLI

    Codex CLI 0.142.1

    1 new

    Added opt-in Windows system proxy support for authentication, including PAC, WPAD, static proxies, and bypass rules.