This week in AI dev tools

Full Changelog : v0.67.0-nightly.20261024.gcd1b45c2d...v0.67.0-nightly.20261025.gcd1b45c2d

Co-authored-by: rekram1-node rekram1-node@users.noreply.github.com

Patch Changes e729d58 : chore(docs): update README.md for consistent format and to avoid duplication Updated dependencies [ 5475347 ] Updated dependencies [ e729d58 ]…

Patch Changes e729d58 : chore(docs): update README.md for consistent format and to avoid duplication Updated dependencies [ 5475347 ] Updated dependencies [ e729d58 ]…

Patch Changes e729d58 : chore(docs): update README.md for consistent format and to avoid duplication Updated dependencies [ a9837eb ] Updated dependencies [ e729d58 ]…

Patch Changes e729d58 : chore(docs): update README.md for consistent format and to avoid duplication Updated dependencies [ 5475347 ] Updated dependencies [ e729d58 ]…

Patch Changes e729d58 : chore(docs): update README.md for consistent format and to avoid duplication Updated dependencies [ 5475347 ] Updated dependencies [ e729d58 ] ai@7.0.153…

Patch Changes e729d58 : chore(docs): update README.md for consistent format and to avoid duplication Updated dependencies [ 5475347 ] Updated dependencies [ e729d58 ]…

Speech Ollama can now generate speech on macOS. ollama run plays text or a chat model's reply aloud as the audio is generated, and apps can stream audio through the…

What's Changed qwen3coder: preserve literal function closing tags in parameters by @hoyyeva in #19228 launch: use allocated context for OpenCode by @hoyyeva in #19193 decision:…

Co-authored-by: github-actions[bot] 41898282+github-actions[bot]@users.noreply.github.com

Fixes and changes

Fixed Managed model "fusion" now starts sessions on Auto and applies fusionTier settings Interactive and SDK Autopilot sessions keep waiting for attached background tasks instead…

What's Changed fix(cli): stop continuation stream processing and auto-nudge after request cancellation by @jesussamuel-byte in #29868 fix(cli): correct Thai and Lao SARA AM…

Changes since 1.2.16 release(langgraph): 1.2.17 ( #9418 ) fix(langgraph): apply an explicit recursion_limit equal to the default ( #9417 ) fix(langgraph): keep the caller's…

The Copilot usage metrics API now reports model usage from more of the tools your developers use with Copilot, including tools like OpenCode that weren't previously captured in…

The new GitHub app in Teams brings notifications and Copilot together, helping teams turn issues into fixes and ship code faster without leaving the conversation. Get notified in…

September 2026 brought deeper visibility into your work: explore pull requests without leaving chat, bring the AI model your team trusts, and fix vulnerable packages the moment…

What's changed Updated to parity with Claude Code v2.1.310 Update npm install @anthropic-ai/claude-agent-sdk@0.3.310 # or yarn add @anthropic-ai/claude-agent-sdk@0.3.310 # or…

What's changed Added the Monitor tool and plugin monitors on Bedrock, Vertex, and Foundry, and when telemetry or nonessential traffic is disabled; long foreground sleeps are now…

Bug Fixes google-cloud: send user profile routes to the organization endpoint Chores internal: stop expecting the beta query param in tests Documentation google-cloud: update the…

Chores internal: stop expecting the beta query param in tests

Features api: add Claude Tag endpoints for Claude Enterprise organizations Bug Fixes client: link the long-request error messages to the SDK documentation Chores client: stop…

Features api: add support for message threads

Features api: add support for message threads

Features api: add support for message threads

Features api: add Claude Tag endpoints for Claude Enterprise organizations Bug Fixes google-cloud: send user profiles requests to the organization endpoint Chores client: stop…

Features api: add support for message threads api: add usage and cost reports by dimension, mark older reports legacy Bug Fixes client: stop suggesting math.inf in the…

Fixes and changes

Improved Assisted Permissions are available without enabling experimental mode. Fixed When a saved Auto preference is no longer available, Auto falls back to its default…

What's Changed Update changelog by @Omar-V2 in #220 Add safety_identifier to chat.create by @rokujix in #221 Add guidelines to chat.create by @rokujix in #229 Allow protobuf 7 by…

Patch Changes Updated dependencies [ 9ec221d ] ai@5.0.281

Patch Changes @ai-sdk/harness@1.0.163

Patch Changes ai@7.0.152

Patch Changes ai@7.0.152

Patch Changes ai@7.0.152

Patch Changes ai@6.0.315

NVIDIA Nemotron 3.5 Bolt E4B Nemotron 3.5 Bolt E4B is NVIDIA’s smaller, Gemma 4-based vision-language model for fast local conversations, image understanding, and tool use. It…

renderers: add gemma4-nemotron renderer Add a Gemma 4 renderer variant that renders assistant content before its tool calls rather than after the tool responses. Models opt in…

Speech Ollama can now generate speech with Phonon on Apple Silicon. ollama run plays text or a chat model's reply aloud as the audio is generated, and apps can stream audio…

Learn how OpenAI scaled Habitat with reliable multi-tenancy, global caching, and Azure Cosmos DB to deliver 12× faster reads for ChatGPT.

OpenAI explains how we strengthened containment for frontier model training and evaluation, including offline environments and high-compute red-teaming.

OpenAI and Poland’s NASK are partnering on AI safety, security, and technical research to strengthen safeguards and support a more resilient digital ecosystem.

See how five builders are putting Gemini 4 Argon to work — and get inspired for your next build.

✨ Features Automatic thinking level operation for improved model reasoning #12237 Resume failed or cancelled delegate subagents #13015 Subagent session link in UI with view-only…

Changes since cli==0.4.37 release(cli): 0.4.38 ( #9411 ) feat(cli): add the priority section to langgraph.json ( #9393 ) chore(deps): bump uv from 0.11.15 to 0.13.4 in /libs/cli…

Changes since sdk==0.4.8 release(sdk-py): 0.4.9 ( #9398 ) feat(sdk-py): add Priority for thread priority settings ( #9392 ) chore(deps): remove precautionary third-party upper…

New Features Save reusable prompts and slash commands in ten numbered /presets slots, with a keyboard menu for quick access. ( #54380 , #54381 ) Set model, reasoning effort,…

Updated dependencies to address recent software ecosystem vulnerabilities. ( #66023 , #65963 )

Updated dependencies to address recent software ecosystem vulnerabilities. ( #66022 , #65962 )

Co-authored-by: roboeve[bot] 311457182+roboeve[bot]@users.noreply.github.com

release: v2.1.4

🔄 Keep Bucket listings fresh with live_follow=True HfFileSystem caches directory listings, so files added to a Bucket by another process did not show up until you called…

What's Changed fix(a2a-server,core): preserve extended token usage metadata and align per-turn reporting by @jvargassanchez-dot in #29834 fix(core): ensure all model turns with…

https://code.visualstudio.com/updates/v1_143

https://code.visualstudio.com/updates/v1_143

Replace the prerelease checkpoint path with Gradium's GA format, including default-voice import and speech conditioning. Set the 0.40.9 minimum, update the speech docs, and test…

flint: fix replies with thinking disabled Close the thinking section when think:false so Flint answers directly. Default and think:true prompts are unchanged. update template

What's Changed llama.cpp update Full Changelog : v0.40.8...v0.40.9-rc0

Copilot usage metrics now show the names of the skills, custom agents, Model Context Protocol (MCP) servers, slash commands, and plugins that your organization creates, instead…

In September, we announced a global default policy for generally available GitHub Copilot features and supported client capabilities on Copilot Business and Copilot Enterprise…

The Copilot impact dashboard now includes an Industry comparison section. Enterprises, and organizations that aren't owned by an enterprise, can compare their Copilot agentic…

What's changed Fixed sandboxed Bash commands on macOS and Linux running the shell's startup files outside the sandbox before each command Fixed macOS sandbox denyWrite glob…

What's changed Bug fixes and reliability improvements

What's changed Fixed host-supplied managedSettings dropping disableAllHooks , disableSkillShellExecution and empty allowedHttpHookUrls / httpHookAllowedEnvVars lists Changed…

What's changed Updated to parity with Claude Code v2.1.308 Update npm install @anthropic-ai/claude-agent-sdk@0.3.308 # or yarn add @anthropic-ai/claude-agent-sdk@0.3.308 # or…

Patch Changes Updated dependencies [ 8202f5b ] Updated dependencies [ 85d84a1 ] @ai-sdk/provider-utils@5.0.67 @ai-sdk/provider@4.0.33 @ai-sdk/openai-compatible@3.0.77

Patch Changes Updated dependencies [ 8202f5b ] Updated dependencies [ 85d84a1 ] @ai-sdk/provider-utils@5.0.67 @ai-sdk/provider@4.0.33

Patch Changes @ai-sdk/harness@1.0.161

Patch Changes Updated dependencies [ 8202f5b ] Updated dependencies [ 8e18d59 ] Updated dependencies [ 85d84a1 ] @ai-sdk/provider-utils@5.0.67 ai@7.0.150 @ai-sdk/provider@4.0.33

Patch Changes Updated dependencies [ 8202f5b ] Updated dependencies [ 8e18d59 ] @ai-sdk/provider-utils@5.0.67 ai@7.0.150

Patch Changes Updated dependencies [ 10026d1 ] @ai-sdk/provider@4.0.32 @ai-sdk/openai-compatible@3.0.76 @ai-sdk/provider-utils@5.0.66

Fast, focused, low-cost decisions for any application.

What forward deployed engineering means at OpenAI

Uber is rolling out ChatGPT Work to more than 30,000 employees, helping teams research, analyze data, create presentations, and complete complex workflows.

Changes since langchain-core==1.6.13 release(core): 1.7.0 ( #41418 ) feat(core): emit tool-call-delta for streamed tool-call args ( #41257 ) Notes Tool-call args now stream as…

2026-10-22 Clicking a completed response code block copies its original code when enabled. ACP sessions now honor saved sandbox settings and --sandbox/--no-sandbox flags,…

Added Clicking a completed response code block copies its original code when enabled. Improved Reuse cached local MCP tool snapshots across registered Git worktrees in one…

Added enable_mouse_capture setting (or VIBE_ENABLE_MOUSE_CAPTURE=false ) to leave text selection and copy to your terminal; depending on the terminal, some other mouse…

v2.3.0.rc1 pre-release

v2.3.0.rc0 pre-release

Core Improvements Upgraded the bundled Bun runtime to 1.4.4, which includes security fixes for Brotli, TLS, HTTP/1.1, HTTP/2, HTTP/3, WebSocket, and fetch , and updates root…

release: v2.1.3

Summary This is my first PR; I'm still learning Rust. We made long project names show dots at the end when they don't fit, so the shortcut stays visible. We made the shortcut go…

What's Changed Changelog for v0.66.0-preview.0 by @gemini-cli-robot in #29852 feat(cli): bridge the IDE companion into container sandboxes over exec stdio by @elberthc-byte in…

What's changed Added wakeup_source to the command_lifecycle started frame of a scheduled wakeup or task, and scheduled_for for a wakeup (capability lifecycle_wakeup_source_v1 ),…

What's changed Updated to parity with Claude Code v2.1.306 Update npm install @anthropic-ai/claude-agent-sdk@0.3.306 # or yarn add @anthropic-ai/claude-agent-sdk@0.3.306 # or…

What's changed Added /claude-api api-version to move Claude API code to API version 2026-10-21 Added the claude_code.safety_event OpenTelemetry event for each tool call a…

What's changed Fixed claude -p inside cloud sessions reporting "Not logged in" when started without the session's environment variables Fixed an API error being replaced by…

Features api: add enforcement status, actions and filter to user profiles

⚠ BREAKING CHANGES api: add support for new API version 2026-10-21 Features api: add a workspace_id filter and the usage_tier scope type to list spend limits api: add support for…

Features api: add enforcement status, actions and filter to user profiles

⚠ BREAKING CHANGES api: add support for new API version 2026-10-21 Features api: add support for new API version 2026-10-21

⚠ BREAKING CHANGES api: add support for new API version 2026-10-21 Features api: add support for new API version 2026-10-21

⚠ BREAKING CHANGES api: add support for new API version 2026-10-21 Features api: add support for new API version 2026-10-21

⚠ BREAKING CHANGES api: add support for new API version 2026-10-21 Features api: add support for new API version 2026-10-21

⚠ BREAKING CHANGES api: add support for new API version 2026-10-21 Features api: add support for new API version 2026-10-21

Recursive Flint Flint brings Recursive's coding model to Ollama. It is designed for the work coding agents do across a repository: understanding unfamiliar code, tracing bugs,…

We were still using a pre-release, and the download site is flaky, so lets get current on the latest CUDA Toolkit for WoA support.

Preserve source tool JSON alongside decoded API values so Flint can retain numeric types and schema order in prompts. Existing API marshaling is unchanged. Assemble ModelOpt…

What's Changed pplx-decider is a 27B decision model from Perplexity, fine-tuned from Qwen3.8-27B Continuing an OpenAI-compatible Responses conversation no longer fails when…

Changes since langchain==1.4.6 chore(langchain): update lockfile ( #41402 ) release(langchain): 1.4.7 ( #41401 ) chore(deps): bump certifi from 2025.11.12 to 2026.10.19 in…

Changes since langchain-core==1.6.12 release(core): 1.6.13 ( #41397 ) chore(deps): bump certifi from 2025.11.12 to 2026.10.19 in /libs/core ( #41373 )…

Changes since langchain-perplexity==1.4.1 release(perplexity): 1.4.2 ( #41392 ) chore(model-profiles): refresh model profile data ( #41354 ) chore(deps): bump certifi from…

How OpenAI works with distribution partners to uphold consistent safeguards, support responsible access, and clarify expectations in our Usage Policies.

AT&T is working with OpenAI to make legal knowledge easier to access. Learn how its LegalEdge team is rethinking intake and scaling legal judgment.

From local businesses to global brands, more advertisers can now reach people as they explore, compare and decide what to buy in ChatGPT.

OpenAI and Google Cloud are bringing frontier models and Codex workflows to Gemini’s Model Garden, available through Google Cloud Marketplace.

Patch the Planet connects security researchers, maintainers, and incident responders to help address critical vulnerabilities across open source.

Learn how OpenAI built ChatGPT’s LLM-powered ad stack to understand conversational intent, keep ads separate from answers, and deliver relevant ads at scale.

This week's release includes Pyrefly as an optional built-in Python language server, the ability to expand and inspect nested variable values directly in debugger hovers, and…

This week's release includes smaller Zed binaries across macOS, Windows, and Linux, the ability to apply inlay hint text edits by double-clicking, context compaction and session…

Fix on-type formatting using a stale position when autoindent is pend…

v1.25.x stable for @JosephTLyons

v1.26.x preview for @JosephTLyons

What's Changed chore: update SDK from production OpenAPI by @github-actions[bot] in #261 Release v3.0.12 by @github-actions[bot] in #262 Full Changelog : v3.0.11...v3.0.12

Co-authored-by: github-actions[bot] 41898282+github-actions[bot]@users.noreply.github.com

What's Changed Fix double border on v0-clone prompt input by @alex-grover in #256 chore: update SDK from production OpenAPI by @github-actions[bot] in #260 Release v3.0.11 by…

2026-10-21 Prompt mode now honors managed Auto model selection for -p and piped stdin runs Subagents honor custom and MCP tool approvals, avoiding duplicate permission prompts…

Fixed Prompt mode now honors managed Auto model selection for -p and piped stdin runs Subagents honor custom and MCP tool approvals, avoiding duplicate permission prompts and…

Improved Sandboxed network traffic now honors NO_PROXY and ALL_PROXY when inheriting host proxy settings Fixed Plugin marketplaces accept git-subdir sources and long plugin…

Mistral is collaborating with NVIDIA to develop a new generation of AI-native engineering workflows powered by NVIDIA accelerated computing.

We just opened a research preview for users on Pro and Max plans: Claude Duet, a new app where Claude works as a proactive agent. For more information, see our blog post:…

fix(app): derive CSS orb motion from the iOS simulation

chore(function): remove exchange_github_app_token_with_pat route ( #57 …

release: v2.1.2

Patch Changes 73a79d7 : Defer validation schema construction until first use to reduce provider import time. Updated dependencies [ d7b0ffa ] Updated dependencies [ 9944649 ]…

Patch Changes 8e40444 : Defer validation schema construction until first use to reduce provider import time. Updated dependencies [ d7b0ffa ] Updated dependencies [ 9944649 ]…

Patch Changes 1a054af : Defer construction of provider validation schemas until first use to reduce package import time. Updated dependencies [ b089ce8 ] Updated dependencies […

Patch Changes 39a776b : Add a shared encodePathSegment helper and use it to encode provider-returned identifiers in credentialed follow-up request paths. 9b368c2 : Defer…

Patch Changes @ai-sdk/harness@1.0.159

Patch Changes Updated dependencies [ b089ce8 ] Updated dependencies [ 39a776b ] @ai-sdk/provider-utils@5.0.65 ai@7.0.148

What's Changed chore: migrate gsutil usage to gcloud storage by @gurusai-voleti in #29839 fix(core): preserve user-level ~/.gitconfig in shell and git environments by…

Features api: add org webhook endpoints and user profile enforcement events Chores ci: run the changelog check on a standard runner tests: make credential tests independent of…

Features aws: require @anthropic-ai/sdk 0.136.0 or later aws: throw an AnthropicError when actionURLs is read on userProfiles or beta.userProfiles , which the AWS gateway doesn't…

Features bedrock: send user_profile_id in the request body on AnthropicBedrockMantle Messages and token counting requests bedrock: throw an AnthropicError when userProfiles or…

Features vertex: throw an AnthropicError when userProfiles or beta.userProfiles is read on AnthropicVertex , which doesn't serve the user profile endpoints

Features google-cloud: use the v1 endpoints and add the organization endpoint

Features api: add data_refreshed_at and has_more to analytics report pages api: add workspace rate-limit updates to the Admin API api: make User Profiles generally available…

Features api: add org webhook endpoints and user profile enforcement events api: add RBAC Group owner-access error codes to the Admin API api: document the unsupported_charset…

Features api: add data_refreshed_at and has_more to analytics report pages api: add workspace rate-limit updates to the Admin API api: make User Profiles generally available…

Minor Changes 7b05470 : feat: Remove the Runloop sandbox provider and its public API; see the migration guide for supported alternatives. Patch Changes 4521cf5 : fix: refresh…

Patch Changes Updated dependencies [ 4521cf5 ] Updated dependencies [ 989513a ] Updated dependencies [ 1163b0c ] Updated dependencies [ 9687b54 ] Updated dependencies [ 7b05470 ]…

Patch Changes 4521cf5 : fix: refresh WebSocket dependencies and development tooling. 9687b54 : fix: Reject WebRTC connections when the server reports an error before the initial…

Patch Changes 989513a : fix: Preserve bounded nested numeric token usage breakdowns across streamed runs, usage accounting, and saved run state (schema 1.23, with backward reads…

Minor Changes 989513a : fix: Preserve bounded nested numeric token usage breakdowns across streamed runs, usage accounting, and saved run state (schema 1.23, with backward reads…

Release v0.21.0

What's Changed fix(cli): enforce read-only workspace settings in untrusted folders by @jvargassanchez-dot in #29583 fix(core): avoid duplicate tool response turns when resuming…

What's Changed Update ROADMAP.md to no longer reference outdated 2025 roadmap by @DavidAPierce in #29723 Changelog for v0.65.0-preview.0 by @gemini-cli-robot in #29739…

What's Changed fix(ui): preserve terminal scrollback during viewport refreshes by @diegogodinezr in #29792 fix(cli): decode streamed Node HTTP activity chunks with StringDecoder…

Deduplicate MLX CUDA runtime payloads Move CUDA libraries shared by the GGML and MLX backends into the sibling CUDA payload, and teach in-process and subprocess MLX loading to…

What's changed Added sandbox.allowNestedUserNamespaces so sandboxed commands on Linux can create user namespaces (Chromium, rootless podman); this lets them get around the…

What's changed Added recommendations and fix to the get_usage response's behavior windows Fixed servers set with setMcpServers being replaced later in the session by a same-named…

2026-10-20 Add native structured output for dynamic workflow subagents on supported models Add assistedApprovalInstructions setting to guide Assisted Permissions confirmation…

Improved Resuming a session shows the conversation sooner without waiting for placeholder environment setup Starting a session with sandboxed MCP servers now avoids unnecessary…

Patch Changes @ai-sdk/harness@1.0.158

Patch Changes Updated dependencies [ 66e74dc ] ai@7.0.147

Patch Changes Updated dependencies [ 66e74dc ] ai@7.0.147

Patch Changes Updated dependencies [ 0c2a9de ] Updated dependencies [ 200c47f ] @ai-sdk/provider@4.0.31 @ai-sdk/provider-utils@5.0.64 @ai-sdk/openai-compatible@3.0.74

Patch Changes Updated dependencies [ 0c2a9de ] Updated dependencies [ 200c47f ] @ai-sdk/provider@4.0.31 @ai-sdk/provider-utils@5.0.64

Patch Changes Updated dependencies [ ca2a9f7 ] @ai-sdk/harness@1.0.157

Co-authored-by: github-actions[bot] 41898282+github-actions[bot]@users.noreply.github.com

We’re announcing new tools for Gemini — including a dedicated business landing page — to help small business owners around the world.

Added New enable_auto_approve setting: when the admin-managed config sets it to false , the auto-approve (YOLO) mode is forbidden — the auto-approve agent (and any custom agent…

release: v2.1.1

release: v2.1.0

Fixed OpenCode models failing to load with a missing field npm error. ( #65875 )

Fixed OpenCode models failing to load with a missing field npm error. ( #65874 ) Added HTML syntax highlighting for svg , mathml and raw tagged templates, and CSS syntax…

An independent Schlotzsky’s and Cinnabon franchisee cut menu repricing time by 95% and built a workflow that could free up $100K in cash flow.

Terminal X’s Eval Agent grades outputs against a 54-criterion rubric, catching 25% more deviances from a fund’s mandate with GPT-6.1 Sol.

OpenAI is acquiring Orchestra Bio to bring AI-powered operations to life sciences R&D, helping biopharma teams accelerate drug development and advance new treatments.

Define an agent as a folder of code, test it with evals, and deploy it to Cursor or your own infrastructure.

What's changed Added a transcript line when a plugin that runs first skips one of your mod's hooks, naming the event and the plugin Added decider: "person" to a mod's tool.check…

What's changed Added suspendSseStream(query) and resumeSseStream(query) to the browser SDK: close a query's event stream and reopen it later without ending the query Added stable…

You can now use the REST API to retrieve and update Copilot model policies for enterprises, enterprise teams, enterprise apps, and organizations. What's new GET and PATCH…

As of today, October 19, 2026, we have deprecated the following models across most GitHub Copilot experiences (including Copilot Chat, inline edits, ask and agent modes, and code…

Added Add native structured output for dynamic workflow subagents on supported models Add assistedApprovalInstructions setting to guide Assisted Permissions confirmation…

2026-10-19 Add --disable-client-telemetry flag to turn off client telemetry uploads from the CLI Add --trust-folder for one-off prompt-mode sessions started with -p or piped…

PR 15095 head at consolidation ( 9338bc9 )

Patch Changes Updated dependencies [ b3c2bb4 ] @ai-sdk/openai-compatible@2.0.86

Patch Changes Updated dependencies [ b3c2bb4 ] @ai-sdk/openai-compatible@2.0.86

Patch Changes Updated dependencies [ 869601f ] @ai-sdk/openai-compatible@3.0.73

Patch Changes @ai-sdk/harness@1.0.156

Patch Changes ai@7.0.145

Patch Changes ai@7.0.145

Patch release: v5.20.2 This patch notably solves an issue where some inputs could loose their original type inside prepare_inputs_for_generation in generate , leading to problems…

Added The app-server reports each model's compaction threshold, so a client can show how full a session's context is before it compacts. Desktop can import projects and sessions…

Here’s how athlete stylist Sydney Bordonaro uses Gemini to build looks for clients — and how you can build your own, too.

What's Changed Add PostgreSQL connection URI support for SQL and key-value stores. #6755 Remove the unused nltk dependency. #6734 Update the UI client lockfile to ogx-client…

Summary I think we are getting closer to fully making PathStyle a mirrored, cross-platform counterpart to Path on top of #65782 . Before the final step — a separate change…

Preserve supported v2 session/update notifications received before session/new reveals the session ID. Register a creation's admission window before publication, stage only…

Added Gemini 4 Argon to the Google AI models ( #65725 )

Changes since cli==0.4.36 release(cli): 0.4.37 ( #9351 ) fix(cli): stop printing subprocess stdin in verbose output ( #9350 ) Merge PR #9349 fix(cli): default deploy to…

Learn how OpenAI is improving ChatGPT across low-resource programming languages, with new evaluations for software engineering, SQLite, and COBOL migration.

Full Changelog : v0.66.0-nightly.20261018.ged96faedf...v0.66.0-nightly.20261019.ged96faedf