Anthropic Releases
Anthropic news, Claude releases and SDK updates. New features, breaking changes, security advisories and deprecations - each summarised in plain English and updated continuously.
Tracking 218 Anthropic releases · Updated
- Claude Code Releases sdkaideprecationengineermediaenergy ·
Claude Code v2.1.212: New /fork behavior, background tasks, and fixes
This update for Claude Code introduces a new behavior for `/fork`, copying conversations to background sessions while users continue working. It also adds features like automatic mode reset, session limits for web searches and subagent spawns, and background processing for long-running MCP tool calls. Numerous bug fixes address issues with plan mode, worktree creation, tool failures, and UI elements, improving overall stability and user experience for engineers interacting with code.
patch - Anthropic Python SDK Releases sdkai ·
Anthropic SDK Python v0.117.0 Adds Dreaming and MCP Tunnels Support
Anthropic's Python SDK has been updated to version 0.117.0, introducing support for the 'dreaming' feature and MCP Tunnels. This release also includes a bug fix to enhance credential security by keeping sensitive material out of traceback frames. The update is relevant for developers using the Python SDK to integrate with Anthropic's AI models, with new features available now.
feature patch - Claude Platform Release Notes aipreview ·
Claude Platform Updates Include New API Features and Model Enhancements
Anthropic has released several updates to its Claude Platform, including new API capabilities for managing organizations and conversations, and enhancements to model features like context windows and memory management. These changes aim to improve developer control, user experience, and model performance for a wide range of users, from individual developers to enterprise organizations. Notable additions include mid-conversation system messages, an Admin API for user management, and a new Sonnet 5 model with expanded capabilities, some of which are currently in beta or preview.
feature patch deprecation announcement - Claude Code Releases sdkaiengineer ·
Claude Code v2.1.211: New flags, numerous bug fixes, and enhancements
This patch release for Claude Code introduces a new flag and environment variable to include subagent thinking in stream-json output. It addresses a wide range of bugs, including issues with permission previews, auto mode, session management, and file upload validation. Enhancements include improved terminal performance, more accurate background agent result reporting, and support for scientific notation in environment variables, benefiting developers and engineers using Claude Code.
patch - Claude Code Releases sdk ·
Claude Code v2.1.210: Elapsed Time Counter, Permission Warnings, and Fixes
Claude Code version 2.1.210 introduces a live elapsed-time counter for long-running tool calls, providing better visibility into progress. It also adds startup warnings for specific permission rules and resolves numerous bugs related to Git operations, plugin behavior, and UI rendering. This update aims to improve the user experience and stability for developers using Claude Code for code generation and editing tasks.
patch - Claude Release Notes gahealthcare ·
Anthropic Claude adds self-serve HIPAA configuration
Anthropic now allows eligible administrators to manage HIPAA readiness for Claude Enterprise and the Claude Platform through a self-serve flow. This enables organizations to review the Business Associate Agreement (BAA), access an implementation guide, and enable HIPAA configuration directly within the platform. This update is relevant for organizations in the healthcare sector needing to ensure compliance with HIPAA regulations when using Claude.
feature announcement - Claude Platform Release Notes aipreviewengineer ·
Claude Enterprise Admin API Beta for User Management
Anthropic is releasing a beta version of the Admin API for Claude Enterprise, enabling organizations to manage users and roles programmatically. This feature allows for detailed control over membership, invitations, and custom roles, which is crucial for enterprise-level administration and security. It is available to all Claude Enterprise organizations with specific beta headers required for certain advanced functions.
feature announcement - Anthropic News aiengineereducation ·
Anthropic Launches Claude for Teachers with Free Premium Access
Anthropic has launched Claude for Teachers, offering verified US K-12 educators free access to premium Claude capabilities, teaching skills, and curriculum connections. This initiative aims to help teachers implement best practices, address time constraints, and improve student outcomes by integrating AI support into instructional planning and differentiation. The service includes integrations with various K-12 tools and is built with a focus on privacy and FERPA compliance. Educators can sign up for free access until June 30, 2027.
feature announcement - Anthropic News aigaengineer ·
Anthropic Invests $10M CAD in Canadian AI Research and Startups
Anthropic is committing $10 million CAD to fund AI research and development at leading Canadian institutions and support startups. This investment aims to advance beneficial and responsible AI applications, building on Canada's historical contributions to AI research. The initiative includes partnerships with major AI institutes and provides Claude API credits to researchers and startups, alongside a report on Claude usage in Canada.
announcement feature - Claude Code Releases sdkaiengineer ·
Claude Agents background sessions fixed
A fix has been deployed to address an issue where models and dialogs were blocked in Claude Agents background sessions. This reverts an overly broad guardrail that was inadvertently causing problems. The update ensures smoother operation for users running background agent tasks.
patch - Claude Code Releases sdkaimedia ·
Claude Code 2.1.208: Screen Reader Mode, Vim Remaps, and Numerous Fixes
Claude Code version 2.1.208 introduces an opt-in screen reader mode for improved accessibility and adds a new setting for Vim insert mode key remapping. This update also includes significant bug fixes across various functionalities, such as session saving, context window handling, and tool execution. These improvements are relevant to all users of Claude Code, particularly those relying on screen readers or advanced Vim configurations.
patch feature - Claude Code Releases sdkaimedia ·
Claude Code v2.1.207: Auto Mode, Bug Fixes, and Security Improvements
Version 2.1.207 of Claude Code introduces auto mode availability on major cloud platforms without opt-in, enhancing user workflow. This release also addresses numerous bugs, including issues with terminal responsiveness, consent dialogs, update processes, and file handling, ensuring a more stable and secure experience. Notably, it fixes security vulnerabilities related to prompt injection and shell injection for plugin developers. These updates benefit all users of Claude Code, particularly those leveraging cloud integrations and developing custom plugins.
patch security - Claude Platform Release Notes aigaengineer ·
Claude Platform Updates: New Sonnet 5, API Key Expiration, Agent Memory
Anthropic has released Claude Sonnet 5 with a 1M token context window and updated pricing, introducing significant behavioral changes and a new tokenizer for Sonnet 5. The Claude Platform now supports API key expiration in the Console, enhancing security for administrators. Additionally, updates to agent memory handling and SDKs improve developer experience and tool integration across various models.
feature patch deprecation - Claude Code Releases sdkaigaengineer ·
Claude Code v2.1.206: Enhanced Commands and Agent Upgrades
Claude Code v2.1.206 introduces directory path suggestions for `/cd`, a `/doctor` check to trim `CLAUDE.md` files, and improved `/commit-push-pr` functionality. These updates enhance developer workflow and agent management, with background agents now upgrading seamlessly post-Claude Code updates. The release also addresses numerous bugs, including login failures and input responsiveness issues across various commands and server configurations.
patch - Claude Platform Release Notes aipreviewengineer ·
Claude API Update: Agent Memory Listing Changes
Anthropic is introducing a new beta header, `agent-memory-2026-07-22`, for the List Memories endpoint, altering response ordering and parameter behavior. This change impacts how developers retrieve memory data and requires updates to SDK usage and explicit beta header configurations. The new header will eventually replace `managed-agents-2026-04-01` on July 22, 2026, and developers are advised to restart pagination when adopting it.
feature breaking announcement - Anthropic News aiengineer ·
UST partners with Anthropic to embed Claude AI in physical product development
UST is integrating Anthropic's Claude AI into its engineering and manufacturing processes to improve physical AI applications. This partnership aims to catch design flaws earlier, speed up validation cycles, and enhance operational efficiency across various industries. UST is also training 20,000 of its engineers on Claude to facilitate AI adoption and deployment within its client solutions.
announcement feature - Claude Platform Release Notes aigaengineer ·
Claude Platform Updates: API Keys, New Models, and Tooling Enhancements
Anthropic has released updates to the Claude Platform, including new model versions like Claude Sonnet 5, Fable 5, and Mythos 5, and enhanced tooling for API keys and managed agents. These changes offer improved context windows, adaptive thinking, and more granular control over agent behavior and security. The updates affect developers building applications with Claude models, providing new capabilities and requiring some migration steps for certain features.
feature patch deprecation announcement - Anthropic News ai ·
Anthropic Seeks Public's Hardest AI Questions for Transparency Initiative
Anthropic is launching an initiative to gather the public's most challenging questions about artificial intelligence, aiming to transparently address societal hopes and concerns. This effort includes large-scale surveys, focus groups, and a public commitment to report on actions taken to mitigate AI risks and maximize benefits. The company, a Public Benefit Corporation, seeks to align AI development with public interest and invites direct input via a dedicated website.
announcement - Anthropic News aigovernancefinance ·
Ben Bernanke Appointed to Anthropic's Long-Term Benefit Trust
Anthropic's Long-Term Benefit Trust (LTBT) has appointed former Federal Reserve Chair Ben Bernanke as a new member to oversee the responsible development of advanced AI for humanity's long-term benefit. His expertise in economics and navigating financial crises will inform Anthropic's approach to AI's economic impacts. Bernanke joins an independent body tasked with ensuring AI development aligns with societal good.
announcement - Claude Release Notes aipreviewengineer ·
Claude Adds Monthly Usage Recap and Focus Tools
Anthropic introduced a new 'Monthly Recap' feature in Claude, providing insights into user activity and work patterns with the AI model. This feature aims to enhance user awareness and productivity by offering data on topics, active days, and peak hours. It is currently in beta for Free, Pro, and Max plan users on web and desktop versions, requiring memory to be enabled, and is complemented by new time management tools.
feature announcement
About Anthropic release tracking on ReleaseBytes
Anthropic ships Claude model releases, API updates and SDK versions on a rapid cadence. ReleaseBytes tracks the official Anthropic announcement channels and SDK changelogs, summarising each release and flagging model deprecations and breaking changes so teams building on Claude stay ahead of them.
Frequently asked questions
How often are Anthropic release notes updated on ReleaseBytes? ›
Continuously. ReleaseBytes monitors the official Anthropic release channels around the clock and publishes a plain-English summary of each announcement shortly after it lands.
What kinds of Anthropic changes does ReleaseBytes track? ›
New features, enhancements, bug fixes, security advisories, breaking changes, deprecations and end-of-life announcements. Every item is tagged by type so you can filter to just the changes that need action.
How can I get alerts for new Anthropic releases? ›
Set up a free email or Slack alert filtered to Anthropic, subscribe to the weekly digest, or follow the RSS feed. Teams can also install the ReleaseBytes GitHub App or connect via MCP.
Where does the Anthropic release data come from? ›
From the official sources: Anthropic news, Claude releases and SDK updates. Every item links back to the original vendor announcement.