Dynamic Workflows for Claude Managed Agents
Enabled dynamic workflows for Claude Managed Agents in beta, allowing agents to create and execute workflows for complex tasks.
Anthropic updates
Last updated · 212 releases tracked since December 2022
Claude updates, release notes and changelog entries from Anthropic, including Claude Code and the Claude Developer Platform. Entries come from Anthropic announcements, the Claude Help Center release notes and the Claude Platform release notes, each dated and linked to its source.
October 2026 at a glance
Anthropic released Models API enhancements including a line field for model grouping, capabilities.server_tools and capabilities.thinking.types.disabled, and launched Claude Haiku 5.5. SDK toolset updates added browser and computer use tools in beta.
Summary of October 2026, generated from the releases listed below.
This month
30
-29 vs last month
Last 12 months
181
Busiest: September 2026
Most shipped
Developer Tools
51 releases
Latest Anthropic releases
Newest first. Each entry links to the official page it came from.
October 202630 releases
Compliance API Enhancements
Compliance API chat endpoints now return chats from unified Claude experience for Claude Enterprise organizations.
Compliance API chat endpoints now return chats from the unified Claude experience
The Compliance API chat endpoints now also return chats from the unified Claude experience, in beta for Claude Enterprise organizations, with your existing Compliance Access Key.
Compliance API Chat Endpoint Update
Compliance API chat endpoints now return chats from unified Claude experience in beta for Claude Enterprise organizations.
SDK Toolset Updates
Python and TypeScript SDKs added browser and computer use tool classes in beta for Claude Platform.
SDK Browser and Computer Use Tool Beta
Python and TypeScript SDKs now include beta classes for browser and computer use tools, allowing custom automation through subclassing and method implementation.
Developer ToolsSource: Claude Platform Release Notes
Prompt Cache Pricing Update
Reduced prompt cache read price for Claude Sonnet 5.5 from $0.20 to $0.10 USD per million tokens.
Prompt cache price reduction for Claude Sonnet 5.5
We've lowered the price of prompt cache reads on Claude Sonnet 5.5 from $0.20 USD to $0.10 USD per million tokens: 0.05x the base input price instead of 0.1x. Cache writes and all other prices are unchanged.
Managed Agents networking updates for web tools
In Claude Managed Agents, a cloud environment with limited networking now also applies its allowed_hosts to the web_search and web_fetch tools. A web_fetch call for a URL on a host that allowed_hosts does not match returns a url_not_allowed error result to the agent. web_search…
Managed Agents Networking Restrictions
Claude Managed Agents now applies allowed_hosts to web_search and web_fetch tools, restricting access to specified hosts in limited networking environments.
Managed Agents environment networking updates
In Claude Managed Agents, environment networking changes apply allowed_hosts to web_search and web_fetch tools.
Launch of Claude Haiku 5.5
We've launched Claude Haiku 5.5 ( claude-haiku-5-5 ), our most capable model tuned for high-volume and latency-sensitive work. It has a 1M token context window, 128k max output tokens, and adaptive thinking with the effort parameter. It's available on the Claude API, Claude in A…
Claude Max and Team plans include monthly API credits
Claude Max and Team plans now include monthly API credits. To learn how to claim them, see API credits for Max and Team plans.
Browser and computer use tool classes added to SDKs
Python and TypeScript SDKs now include classes for browser use tool and computer use tool (beta).
API Credits for Max and Team Plans
Claude Max and Team plans now include monthly API credits for users.
API Credit Additions
Claude Max and Team plans now include monthly API credits for users.
Models API updates with server tools capabilities
We've added capabilities.server_tools to the Models API. GET /v1/models and GET /v1/models/{model_id} now report whether each model accepts the web search and code execution tools. To check whether a model accepts the code execution tool, read capabilities.server_tools.code_exec…
Models API Enhancements
Added capabilities.server_tools field to Models API to indicate model tool support.
Models API Capability Enhancements
Added capabilities.server_tools and capabilities.thinking.types.disabled fields to the Models API for checking model capabilities.
Models API Capabilities Update
Added capabilities.server_tools field to Models API to indicate model support for web search and code execution tools.
Capabilities.server_tools added to Models API
Added capabilities.server_tools to Models API. Reports whether models accept web search and code execution tools.
Models API Thinking Capabilities
Added capabilities.thinking.types.disabled field to Models API to indicate model support for disabling thinking.
Models API Enhancements
Added capabilities.thinking.types.disabled field to Models API to indicate thinking disable support.
Models API adds thinking disabled capability
We've added capabilities.thinking.types.disabled to the Models API. GET /v1/models and GET /v1/models/{model_id} now report whether each model accepts thinking: {type: "disabled"}, which turns thinking off.
Added capabilities.thinking.types.disabled to Models API
Added capabilities.thinking.types.disabled to the Models API. GET /v1/models and GET /v1/models/{model_id} now report whether each model accepts thinking: {type: "disabled"}, which turns thinking off.
Models API Line Field Addition
Added line field to Models API for model line identification in API responses
Models API Enhancements
Added line field to Models API to indicate model line affiliation (e.g., 'opus' for Claude Opus models).
Models API adds line field for model grouping
We've added a line field to the Models API. GET /v1/models and GET /v1/models/{model_id} now return the model line each model belongs to. Claude Opus 4.5 and Claude Opus 4.6 both report opus, for example. Use line to group models without parsing their IDs. line is null for a mod…
Dreams research preview expanded
Dreams (research preview) now supports Claude Opus 5.5, Claude Fable 5.1, and Claude Sonnet 5.5.
Added line field to Models API
Added a line field to the Models API. GET /v1/models and GET /v1/models/{model_id} now return the model line each model belongs to. Claude Opus 4.5 and Claude Opus 4.6 both report opus, for example.
September 202659 releases
SDKs updates and deprecation of Claude Sonnet 4.5
Admin API methods for organization info, members, invites, workspaces, API keys, rate limits, service accounts, workload identity federation, customer-managed encryption keys, and compliance settings are out of beta. Deprecation of Claude Sonnet 4.5 with retirement on November 3…
Claude Sonnet 4.5 Deprecation and Sonnet 5.5 Migration
Deprecation of Claude Sonnet 4.5 model (claude-sonnet-4-5-20250929) with retirement on Claude API scheduled for November 30, 2026. Recommended migration to Claude Sonnet 5.5.
Admin API updates for multiple SDKs and ant CLI
In Python SDK 1.10.0, TypeScript SDK 0.130.0, C# SDK 12.52.0, Go SDK 1.77.0, Java SDK 2.67.0, PHP SDK 0.53.0, Ruby SDK 1.75.0, and version 1.37.0 of the ant CLI, the Admin API methods for organization info, members, invites, workspaces, API keys, rate limits, service accounts, w…
Developer ToolsSource: Claude Platform Release Notes
Admin API methods move out of beta
Admin API methods for organization management move out of beta. Available under client.organization and ant organization.
Admin API General Availability
Admin API methods for organization management moved out of beta for multiple SDKs and CLI tools.
Launched Claude Sonnet 5.5
Launched Claude Sonnet 5.5 (claude-sonnet-5-5). Available on the Claude API, Claude in Amazon Bedrock, Claude Platform on AWS, Claude on Google Cloud, and Claude in Microsoft Foundry.
Claude Sonnet 5.5 launch
Launched Claude Sonnet 5.5, a faster, lower-cost complement to Claude Opus 5.5 in the Claude 5.5 family.
Build plugins for Claude
You can now submit plugins to the Claude directory through a new developer portal, track them through review, and see usage analytics once they’re live.
Compliance API local session endpoints out of beta
The Compliance API local session endpoints are out of beta for Claude for Microsoft 365 sessions in Excel, PowerPoint, Word, and Outlook ( product_surface values beginning with office_agents ).
Compliance API Enhancements
Compliance API local session endpoints now support Office Agents and transcript retrieval.
Compliance API Activity Feed updates
The Compliance API Activity Feed no longer returns file names, project document names, or artifact titles. The filename and title fields on file, project document, and artifact activities are now always empty or omitted, including on activities recorded before this change. To lo…
Compliance API Activity Feed Update
Compliance API activity feed changes to remove file names and document titles from activity objects
Claude API Cache Diagnostics & Compliance Enhancements
Cache diagnostics feature moved out of beta on Claude API with new diagnostics object implementation. Compliance API local session endpoints expanded to include Chrome sessions.
Billing updates for API refusals
We're resuming billing for refusals that arrive before any output when stop_details.category is "bio", "frontier_llm", or "reasoning_extraction", the categories where we measure low volumes of false positives. Mid-stream refusals were already billed. Refusals billed under this c…
Billing resumes for some refusals; Compliance API updates
Anthropic resumed billing for refusals that arrive before any output in the bio, frontier_llm and reasoning_extraction categories, on all platforms. The Compliance API local session endpoints for Claude for Microsoft 365 sessions are out of beta.
Billing changes for specific refusal categories
Resumed billing for refusals that arrive before any output when stop_details.category is "bio", "frontier_llm", or "reasoning_extraction". Refusals billed under this change are charged like any other request.
Billing changes for refusals and Compliance API updates
Resuming billing for refusals that arrive before any output when stop_details.category is "bio", "frontier_llm", or "reasoning_extraction". Compliance API local session endpoints are out of beta for Claude for Microsoft 365 sessions.
Billing and Compliance API Updates
Billing changes for refusals and Compliance API updates affecting Claude for Microsoft 365 sessions.
Cache Diagnostics Update
Cache diagnostics beta removal and new diagnostics field implementation on Claude API.
Cache Diagnostics General Availability
Cache diagnostics feature moved out of beta with new diagnostics object requirements
Launched Claude Opus 5.5
Launched Claude Opus 5.5 (claude-opus-5-5), a model for long-running agentic coding and knowledge work. 1M token context window, 128k max output tokens, and always-on adaptive thinking.
Claude Opus 5.5 Launch
Launch of Claude Opus 5.5 (claude-opus-5-5) for long-running agentic coding and knowledge work. Features 1M token context window, 128k max output tokens, and adaptive thinking. Available on multiple platforms.
Managed Agents permission policies updated with auto evaluation
Claude Managed Agents permission policies now include auto: the server evaluates each agent or MCP tool call and runs it, denies it, or pauses for your approval. agent.tool_use and agent.mcp_tool_use events report how each call was evaluated in an evaluation field alongside eval…
Compliance API Session Updates
Compliance API local session endpoints now support Chrome sessions for Claude Enterprise organizations.
Compliance API session transcript expansion
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.
Compliance API Enhancements
Compliance API local session endpoint updates and activity feed changes for Claude for Microsoft 365.
Claude API Cache Diagnostics Update & Compliance Changes
Cache diagnostics response format updated to always include diagnostics field. Compliance API Activity Feed now omits file names and titles for privacy compliance.
Cache diagnostics field updates
For cache diagnostics, a response to a request that sends the cache-diagnosis-2026-04-07 beta header now always includes the diagnostics field. The field is null when the request did not include the diagnostics object. Previously the field was omitted in that case.
Cache diagnostics changes
Cache diagnostics changes: diagnostics field always included in responses. Fingerprint stored only when diagnostics object is included.
Create designs, decks, and docs in any conversation
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 a…
Claude Cowork comes to every conversation
Rolled out Claude Cowork experience making all features available in any conversation with cross-platform support for Pro/Max plans.
Salesforce in Claude (beta)
Launched Salesforce plugin with 37 pre-built sales skills for managing accounts, opportunities, and pipeline through Claude.
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 a…
Messages API introduces conversation compaction
The Messages API can now compact a conversation on demand on the Claude API, in beta with the compact-2026-09-04 beta header. Send the top-level compaction parameter, and the API returns a signed compaction block that summarizes the messages you sent. On later requests, send tha…
Messages API Compaction Feature
Introduced conversation compaction feature for Claude API to reduce token usage with beta header.
Messages API Compaction Beta
Messages API compaction beta with compact-2026-09-04 header and thinking-binding-controls-2026-08-01 beta features.
Messages API compact feature and thinking-binding-controls
Messages API can compact a conversation on demand with the compact-2026-09-04 beta header. Thinking-binding-controls beta header added.
Claude API Conversation Compaction & Tool Enhancements
Messages API introduced conversation compaction feature via beta header. Added support for mid-conversation tool definitions and MCP toolset integration.
Smart reports (beta)
Introduced smart reports analyzing team Claude usage patterns, costs, and friction points for Enterprise plans.
InfrastructureSource: Claude Help Center Release Notes
Managed Agents Policy Update
Managed Agents permission policies update with auto evaluation for tool calls and ant CLI session connectivity.
Managed Agents permission policies updated with 'auto' mode
Claude Managed Agents permission policies now include 'auto' mode for server evaluation of agent/tool calls.
Managed Agents permission policies and ant CLI updates
Managed Agents permission policies now include auto. ant beta:sessions connect command added for Claude Managed Agents sessions.
Managed Agents Permission Policies
Updated permission policies for Claude Managed Agents with new networking restrictions and tool usage controls.
Managed Agents Auto-Evaluation Policy
Claude Managed Agents permission policies now include auto-evaluation for agent and MCP tool calls, with detailed evaluation reporting.
Managed Agents auto permission policy and ant CLI session connect
Claude Managed Agents permission policies now include auto evaluation for agent/tool calls
Managed Agents Auto Evaluation Mode
Managed Agents permission policies updated to include auto evaluation mode for tool calls
ant CLI updates with session connection feature
Version 1.32.0 of the ant CLI adds ant beta:sessions connect, which attaches your terminal to a Claude Managed Agents session. You can follow the session live, send messages, and allow or deny tool calls that are waiting for approval. Pass --web to serve the Claude Console's ses…
Developer ToolsSource: Claude Platform Release Notes
ant CLI session connection feature added
ant CLI version 1.32.0 adds 'ant beta:sessions connect' for terminal access to Managed Agents sessions.
Cache diagnostics fingerprint updates
For cache diagnostics, the API now stores a request's fingerprint only when the request includes the diagnostics object. A request that sends only the cache-diagnosis-2026-04-07 beta header is still accepted, but no fingerprint is stored. A later turn that points previous_messag…
Cache Diagnostics Fingerprint Update
Cache diagnostics fingerprint changes and request fingerprint storage updates for Claude API.
Cache diagnostics fingerprint storage change
For cache diagnostics, the API now stores a request's fingerprint only when the request includes the diagnostics object. A request that sends only the cache-diagnosis-2026-04-07 beta header is still accepted, but no fingerprint is stored.
Cache diagnostics fingerprint changes
Cache diagnostics fingerprint changes: fingerprint stored only when diagnostics object is included.
ant CLI Version 1.30.0 Release
ant CLI version 1.30.0 with apply command for resource management and deployment.
ant CLI v1.30.0 adds ant apply
ant CLI v1.30.0 adds ant apply for managing resources as code
ant CLI Resource Management Update
New ant CLI commands for managing resources as code with repository integration
ant CLI apply command
ant CLI apply command added for managing resources as code with claude-lock.json lockfile.
ant CLI adds apply command for resource management
Version 1.30.0 of the ant CLI adds ant apply, which creates and updates agents, environments, skills, memory stores, and deployments from files in your repository. Describe each resource in a file, run ant apply, and approve the plan it prints. Commit the claude-lock.json lockfi…
Developer ToolsSource: Claude Platform Release Notes
Launched Claude Fable 5.1 and Mythos 5.1
Launched Claude Fable 5.1 (claude-fable-5-1) and Claude Mythos 5.1 (claude-mythos-5-1). 1M token context window, 128k max output tokens, and always-on adaptive thinking.
Claude Fable 5.1 and Mythos 5.1 launch
Launched Claude Fable 5.1 and Claude Mythos 5.1, world's most advanced models for coding and knowledge work.
August 202613 releases
SDK Version Updates
SDK updates with beta header removal and new features for file and skill management across multiple programming languages.
Personal and service account API keys in Console
Claude Console can now create personal keys and service account keys that act as a user or service account, can be scoped to a workspace or admin endpoints, and stop working when the linked account is removed. Several SDKs also dropped beta headers for the Files and Skills APIs.
Compliance API session endpoints out of beta; Admin API in SDKs
Compliance API session endpoints for Cowork and Claude Code sessions are out of beta. Admin API available in CLI and multiple SDKs under client.beta.organization.
Memory in Claude Cowork, editable topics, and sensitive topics setting
Introduced memory system with categorized entries, topic management, and sensitive topic settings for Claude Cowork.
InfrastructureSource: Claude Help Center Release Notes
Python SDK v1.0 Release and Toolset Updates
Released v1.0 of Python SDK with HTTP layer change to httpx2. Computer use and browser use toolsets available on Google Cloud for specific Claude models.
Computer use, Files API and Skills API out of beta; browser use tool launched
The computer use tool is out of beta as computer_toolset_20260801, and a new browser use toolset was launched. The Files API and Agent Skills / Skills API are out of beta, and Admin API user-management endpoints for Claude Enterprise are out of beta. Managed Agents gained web search/fetch domain restrictions and a redesigned Console session viewer.
Workbench Renamed to Playground
Workbench is now playground in the Claude Console. Playground supports every Messages API parameter and includes templates that demonstrate API features such as code execution and web search. It shows the full SDK request and the API response for each run, to help you understand…
Compliance API Enhancements
The Compliance API now returns transcripts of Cowork and Claude Code sessions that run on your users' machines, in beta for Claude Enterprise organizations. New endpoints added for session management and transcript retrieval.
Pricing Update for Claude Sonnet 5
Introductory pricing for Claude Sonnet 5 becomes standard price. Scheduled price increase on September 1, 2026 is canceled.
Claude Managed Agents New Features
Claude Managed Agents sessions now support budget controls, advisor models, and data residency settings. New features for session management and resource control.
Skill and plugin security scanning (beta)
Enabled skill and plugin security scanning for Enterprise plans to detect malicious content in third-party assets.
Enterprise & SecuritySource: Claude Help Center Release Notes
Inference hooks beta for Claude Enterprise
Inference hooks are in beta for Claude Enterprise organizations. Claude holds each governed prompt across claude.ai, Cowork and Claude Code for an AI security server's allow or deny verdict before inference proceeds, and denials are recorded in the compliance Activity Feed.
Compliance API for Cowork Sessions
Compliance API now returns transcripts of Cowork sessions started on claude.ai web or mobile. New endpoints for cloud session management.
July 202612 releases
Claude Opus 5 launch
Launched Claude Opus 5, a cost-effective model matching Claude Fable 5 capabilities at half the price.
Claude Managed Agents Enhancements
Claude Managed Agents updates include effort level configuration, webhook enhancements, session initialization, and event delta support for thread streams.
HIPAA configuration for your Claude organizations is now self-serve
You can now manage HIPAA readiness for your Claude organizations yourself. This applies to both Claude Enterprise and the Claude Platform (API). In each product, an eligible admin can review the Business Associate Agreement (BAA), download the implementation guide, and enable th…
HIPAA configuration for your Claude organizations
Enabled self-serve HIPAA configuration for Claude organizations with Business Associate Agreement management and implementation guides.
Enterprise & SecuritySource: Claude Help Center Release Notes
Admin API user management for Claude Enterprise
Admin API beta enables managing Claude Enterprise organization members, groups, and custom roles with specific beta headers and permissions.
Updated memory for Claude
Updated memory system with individual categorized entries and improved context management during conversations.
InfrastructureSource: Claude Help Center Release Notes
A new way to reflect with Claude
Your monthly recap is a new feature at Settings > Reflect that shows you the topics you spent time on, your most active day and peak hour, and observations about how you work with Claude. It's in beta on Free, Pro, and Max plans on the web and Claude Desktop, and requires memory…
API Key Expiration Management
API key expiration settings added to Claude Console with email notifications for keys expiring in 7+ days.
Write tools for the Microsoft 365 connector
The Microsoft 365 connector now goes beyond search. With write tools enabled, Claude can draft, send, and organize email, manage calendar events, update mailbox settings, and create and update files in OneDrive and SharePoint. Read and search tools work as before, and Teams rema…
Claude Cowork on web and mobile
Claude Cowork is now available on web and mobile in addition to desktop. We are rolling this capability out over the next several weeks starting with the Max plan, with more plans to follow. Cowork runs your sessions remotely (in beta), so your sessions and files are saved to yo…
Write tools for the Microsoft 365 connector
The Microsoft 365 connector now goes beyond search. With write tools enabled, Claude can draft, send, and organize email, manage calendar events, update mailbox settings, and create and update files in OneDrive and SharePoint. Read and search tools work as before, and Teams rema…
Model entitlements for Enterprise plans (beta)
Admins on Enterprise plans can now control what models their users have access to, and what effort level settings they can use.
June 20269 releases
Claude Sonnet 5 launch
Claude Sonnet 5 launch with improvements in reasoning, tool use, coding, and knowledge work
Raised rate limits across the 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 lo…
Enable Trusted Devices for Remote Control
Admins for Team and Enterprise plans can require device verification for remote Claude Code sessions
Delegate work to Claude in Slack
Team and Enterprise plans can tag Claude directly in Slack conversations to delegate tasks
Edit Claude’s drafts in place
Edit Claude’s drafts in place When Claude writes a draft in chat or Claude Cowork (a report, a plan, a brief), it opens right beside your chat. Highlight the part you want changed, type the change, and Claude makes the edit right where you marked it. No switching apps, no re-des…
Edit Claude’s drafts in place
When Claude writes a draft in chat or Claude Cowork (a report, a plan, a brief), it opens right beside your chat. Highlight the part you want changed, type the change, and Claude makes the edit right where you marked it. No switching apps, no re-describing paragraphs in chat.
Claude Fable 5 launch
Claude Fable 5 launch as a Mythos-class model safe for general use
Enterprise plans can manage admin permissions with custom roles
Enterprise plans can manage admin permissions with custom roles
Advisor tool output cap and refusal billing change
The advisor tool now supports a max_tokens parameter to cap advisor output per call. On the Claude API, requests that return a refusal before any output is generated are no longer billed.
May 202612 releases
Claude Managed Agents on Claude Platform on AWS
Claude Managed Agents webhooks, multiagent orchestration and self-hosted sandboxes are now available on Claude Platform on AWS.
Claude Opus 4.8 launch
Claude Opus 4.8 upgrade with improvements in coding, agentic skills, and practical knowledge work
Thinking token breakdown in Messages API usage
The Messages API response now includes usage.output_tokens_details.thinking_tokens, reporting how many billed output tokens were extended thinking. No beta header is required.
Enterprise plans can manage connector access with custom roles
We added connector permissions to extend the existing custom roles framework and allow administrators to control which connectors, and which individual tools on those connectors, are available to each custom role.
Claude now works with more security and compliance tools
Claude Compliance API integrations with security and compliance tools
MCP Tunnels and Self-Hosted Sandboxes for Managed Agents
MCP tunnels is now available as a research preview, so you can connect to MCP servers in your private network. Self-hosted sandboxes are now available for Claude Managed Agents, as an alternative to running tool execution in Anthropic's infrastructure. See Self-hosted sandboxes.…
Enhanced Web Search Tool for SEC Filings
The web search tool now returns richer SEC filing data, making it easier to ground financial research agents, earnings analysis, and due-diligence workflows in primary sources with citations.
Cache Diagnostics Beta Launch
We've launched cache diagnostics in public beta. Pass diagnostics.previous_message_id on a Messages request and the API reports a cache_miss_reason explaining where the prompt cache prefix diverged from the previous turn. Include the cache-diagnosis-2026-04-07 beta header in you…
Fast Mode Support for Claude Opus 4.7
Fast mode (research preview) now supports Claude Opus 4.7. Set speed: "fast" with model: "claude-opus-4-7" and the fast-mode-2026-02-01 beta header for significantly faster output token generation at premium pricing. Pricing, rate limits, and access are the same as for Opus 4.6…
Claude Platform on AWS Launch
We've launched Claude Platform on AWS, bringing the Claude API to Anthropic-managed infrastructure accessible through AWS, with AWS billing and IAM authentication. Access the full Messages API, Files API, Message Batches API, Claude Managed Agents, Agent Skills, code execution,…
Multiagent Orchestration and Managed Agents Enhancements
Multiagent orchestration and Outcomes are now in public beta under the standard managed-agents-2026-04-01 beta header. Claude Managed Agents vault credential background refresh is now supported for mcp_oauth credentials. See Authenticate with vaults. Webhooks for Claude Managed…
Workload Identity Federation Launch
Authenticated workloads to the Claude API with short-lived OIDC tokens from your own identity provider.
April 202612 releases
Claude API Skill Release
Released Claude API skill for building on Messages API and Managed Agents
Rate Limits API Launch
Programmatically query organization rate limits via new API
Managed Agents Memory Beta
Public beta for memory feature in Claude Managed Agents
Claude Design by Anthropic Labs
Claude Design launched as an Anthropic Labs product for visual collaboration
Claude Opus 4.7 launch
Claude Opus 4.7 launch with improvements in software engineering and vision capabilities
Role-based access controls for Enterprise plans
Admins can now organize users into groups, manually or via SCIM from your identity provider, and assign each group a custom role defining which Claude capabilities its members can use. Turn Claude Cowork on for specific teams, restrict features by department and adjust as adopti…
Claude Cowork generally available
Claude Cowork generally available with new analytics and OpenTelemetry support
Advisor Tool Beta Launch
Public beta for advisor tool that pairs executor and advisor models
Claude Managed Agents public beta
Public beta for Claude Managed Agents with secure sandboxing and streaming
ant CLI Release
Launched ant CLI for faster Claude API interaction and resource management
Messages API on Amazon Bedrock
Messages API available on Amazon Bedrock as research preview
Claude Mythos Preview research preview
Claude Mythos Preview is available as a gated, invitation-only research preview for defensive cybersecurity work as part of Project Glasswing.
March 202613 releases
Message Batches API Max Tokens Increase
Raised max_tokens cap to 300k on Message Batches API for Opus 4.6/Sonnet 4.6
Interactive apps in Claude for iOS and Android
The Claude mobile app can now connect to fully interactive apps. Pull up live charts, sketch diagrams, and build shareable assets, all rendered visually right in your conversation.
Computer use research preview in Cowork and Claude Code + Dispatch improvements
Users on Pro and Max plans can give Claude access to computer use. Claude can open files, run dev tools, point, click, and navigate to what’s on your screen to perform tasks itself-with no setup required. Claude’s new computer use capability makes Dispatch even more helpful by a…
Models API Capability Fields Addition
Added model capability fields to Models API for better feature discovery
Control Cowork from your phone with a persistent thread (research preview for Pro/Max)
Users on Pro and Max plans can access a persistent agent thread via Claude Desktop or Claude for iOS/Android to manage tasks in Cowork. We're rolling this out to Max plans starting today, and Pro plans over the next two days.
Control Cowork from your phone with a persistent thread
Claude added a persistent Cowork thread for Pro and Max users so tasks can continue from mobile or desktop without losing context.
Display Field for Extended Thinking
Launched display field for extended thinking, allowing omission of thinking content from responses for faster streaming. Set thinking.display: "omitted" to receive thinking blocks with an empty thinking field and preserved signature for multi-turn continuity. Billing remains unc…
1M Token Context Window Launch
1M token context window launched for Claude Opus 4.6 and Sonnet 4.6 at standard pricing. Requests over 200k tokens work automatically without beta headers. Media limit increased to 600 images/PDFs per request with 1M context window.
Claude Partner Network Launch with $100 Million Investment
Anthropic launches the Claude Partner Network with a $100 million investment to support enterprise adoption of Claude, including training, technical certification, and joint market development.
Claude creates interactive charts, diagrams and visualizations
Claude can now create custom charts, diagrams and other visualizations in-line in its responses.
Work across Excel and PowerPoint
We’ve improved our Claude for Excel and Claude for PowerPoint add-ins. They can now share the full context of your conversation, so every action Claude takes in one application is informed by everything that’s happened in the other. We also added support for skills in the add-in…
Partnering with Mozilla to improve Firefox’s security
Collaboration with Mozilla to identify 22 vulnerabilities in Firefox using Claude Opus 4.6, including 14 high-severity issues.
ResearchSource: Anthropic
Memory for free users
Claude made chat memory available to free users so prior context can carry into new conversations without a paid plan.
February 202610 releases
Scheduled tasks in Cowork
Claude introduced recurring and on-demand task scheduling in Cowork, plus a new Customize area in Claude Desktop for skills, plugins, and connectors.
Cowork plugins and admin controls
Claude added a plugin marketplace and admin controls for Team and Enterprise plans, giving workspace owners more structure around Cowork extensions.
Automatic Caching for Messages API
Launched automatic caching for Messages API. Add cache_control field to automatically cache last cacheable block, moving cache point forward as conversations grow. Works with existing block-level cache control for optimization.
Introducing Claude Sonnet 4.6
Claude Sonnet 4.6 is the most capable Sonnet model yet, featuring a 1M token context window in beta, improved coding skills, and enhanced computer use capabilities. It's now the default model for Free and Pro plans with the same pricing as Sonnet 4.5.
Analytics API for Enterprise plans
Anthropic added an Analytics API that lets Enterprise customers pull daily usage and engagement data for Claude and Claude Code Remote programmatically.
Self-serve Enterprise plans
Anthropic opened Enterprise plans for direct purchase, giving organizations a self-serve path to Claude, Claude Code, and Cowork without a sales-led rollout.
Enterprise & SecuritySource: Claude Help Center Release Notes
Fast Mode for Opus 4.6
Launched fast mode in research preview for Opus 4.6, providing 2.5x faster output token generation through speed parameter. Available at premium pricing with waitlist access.
Introducing Claude for PowerPoint
Claude became available as a PowerPoint add-in, extending the product into presentation workflows for paid users.
Claude Opus 4.6 Release
Introducing Claude Opus 4.6 with improved coding skills, longer agentic task support, 1M token context window, and enhanced performance in financial analysis, research, and document creation.
Claude for Excel improvements
Claude for Excel was updated to use Opus 4.6 and to support native spreadsheet actions like pivot table editing and conditional formatting.
January 20265 releases
Structured outputs generally available
Anthropic made structured outputs generally available on the Claude API for Sonnet 4.5, Opus 4.5, and Haiku 4.5. The update expanded schema support, reduced compilation latency, and removed the beta-header requirement.
Claude's new constitution
Publication of the new constitution for Claude, detailing its values, behavior, and training guidelines.
Cowork expansion and Claude Code seat access
Claude expanded Cowork to Pro plan users and added Claude Code access to Team plan Standard seats. The same update also deprecated Opus 4 and 4.1 from Claude and Claude Code.
Cowork research preview on Claude Desktop
Claude launched Cowork as a research preview for Max plans on Claude Desktop for macOS. The feature brought Claude Code's agentic capabilities into the desktop app for knowledge-work workflows.
Advancing Claude in healthcare and the life sciences
Introducing Claude for Healthcare with HIPAA-ready tools and new life sciences capabilities, including improved medical benchmark performance and agent skills.
December 20251 release
Introducing Anthropic Interviewer
Introducing a new tool, Anthropic Interviewer, designed to understand people's perspectives on AI through automated interviews. The tool was tested with 1,250 professionals across different fields, revealing insights into AI's societal and economic impacts.
ProductSource: Anthropic
November 20252 releases
Introducing Claude Opus 4.5
Our newest model, Claude Opus 4.5, is available today. It’s intelligent, efficient, and the best model in the world for coding, agents, and computer use.
Political Even-Handedness Evaluation Methodology Release
Announced a new automated evaluation method to test for political even-handedness in Claude, including open-sourcing the methodology for AI developers to reproduce findings and improve measures of political bias.
ResearchSource: Anthropic
Earlier releases (34)
October 20253 releases
Advancing Claude for Financial Services
Released Claude for Excel in beta with real-time financial modeling capabilities, added connectors for live market data, and introduced new Agent Skills for financial tasks like DCF models and coverage reports
Claude for Life Sciences Release
Announces enhanced capabilities for Claude in life sciences, including new connectors, Agent Skills, and partnerships with scientific platforms.
Introducing Claude Haiku 4.5
Claude Haiku 4.5, our latest small model, is available today to all users. It offers similar coding performance to Claude Sonnet 4 at one-third the cost and more than twice the speed, with improved responsiveness for AI-assisted development workflows.
September 20253 releases
Introducing Claude Sonnet 4.5
Introduces Claude Sonnet 4.5, the most capable coding model with improvements in reasoning, math, and computer use. Includes upgrades to Claude Code, API, and product features like checkpoints and VS Code extension.
Claude Sonnet 4.5 Launch
Introducing Claude Sonnet 4.5, the most capable coding model with improvements in reasoning, math, and computer use. Includes upgrades to Claude Code, API, and apps, along with the Claude Agent SDK for developers.
Claude Sonnet 4.5
Introducing Claude Sonnet 4.5, the most capable coding model with significant gains in reasoning, math, and computer use. Includes major product upgrades like checkpoints, VS Code extension, and context editing features.
August 20254 releases
Updates to Consumer Terms and Privacy Policy
Updates to consumer terms and privacy policy allowing user data to be used for improving Claude models and extending data retention to five years for users who opt in.
ProductSource: Anthropic
Usage Policy Update
Updated Usage Policy to address cybersecurity, agentic use, political content restrictions, and high-risk consumer-facing use cases.
ProductSource: Anthropic
Expanded Claude Access Across U.S. Government Branches
Anthropic expands Claude access to all three branches of the U.S. government, offering Claude for Enterprise and Government at $1, enabling federal agencies to leverage advanced AI capabilities for mission-critical tasks.
Claude Opus 4.1
Claude Opus 4.1 is an upgrade to Claude Opus 4 on agentic tasks, real-world coding, and reasoning. It improves coding performance to 74.5% on SWE-bench Verified, enhances research and data analysis skills, and includes notable performance gains in multi-file code refactoring.
May 20251 release
Introducing Claude 4
Introducing Claude Opus 4 and Claude Sonnet 4, setting new standards for coding, advanced reasoning, and AI agents. Includes extended thinking with tool use, new model capabilities, and Claude Code general availability.
February 20254 releases
Claude's extended thinking
Announces Claude 3.7 Sonnet with extended thinking mode, visible thought process, and improved agentic capabilities for tasks like computer interaction and Pokémon gameplay.
Claude 3.7 Sonnet and Claude Code
Announced Claude 3.7 Sonnet, a hybrid reasoning model with extended thinking mode for complex tasks, and Claude Code, a CLI tool for agentic coding. Available on all plans, Amazon Bedrock, and Vertex AI.
Introducing the Anthropic Economic Index
Announces the launch of the Anthropic Economic Index, an initiative to analyze AI's impact on labor markets and the economy using real-world data from Claude.ai. Includes open-sourcing of datasets for research purposes.
ResearchSource: Anthropic
Constitutional Classifiers: Defending against universal jailbreaks
A new paper from the Anthropic Safeguards Research Team describes a method that defends AI models against universal jailbreaks.
ResearchSource: Anthropic
November 20244 releases
Model Context Protocol Release
Open-sourcing the Model Context Protocol (MCP), a standard for connecting AI assistants to data systems.
InfrastructureSource: Anthropic
Model Context Protocol Launch
Open-sourced Model Context Protocol (MCP) for connecting AI assistants to data systems, including enterprise tools and development environments. Includes specification, SDKs, local server support in Claude Desktop, and pre-built connectors for Google Drive, Slack, GitHub, and ot…
InfrastructureSource: Anthropic
Model Context Protocol
Open-sourcing the Model Context Protocol (MCP), a new standard for connecting AI assistants to data systems like repositories, tools, and environments.
InfrastructureSource: Anthropic
Introducing the Model Context Protocol
Open-sourcing the Model Context Protocol (MCP), a new standard for connecting AI assistants to data systems like repositories, tools, and environments.
InfrastructureSource: Anthropic
October 20242 releases
Claude 3.5 Sonnet on GitHub Copilot
Starting today, Claude 3.5 Sonnet begins rolling out on GitHub Copilot, enabling developers to choose Claude 3.5 Sonnet for coding directly in Visual Studio Code and GitHub.com. The model outperforms others on SWE-bench Verified and achieves top scores on HumanEval.
Claude 3.5 Sonnet
Upgraded Claude 3.5 Sonnet with industry-leading software engineering skills and improved coding performance on benchmarks.
August 20241 release
Claude is now available in Brazil
Claude, Anthropic’s AI assistant, is now available in Brazil via Claude.ai, mobile apps, and API. Offers free and paid plans with access to Claude 3.5 Sonnet and other models.
June 20241 release
Collaborate with Claude on Projects
Claude.ai Pro and Team users can now organize chats into Projects, enabling curated knowledge sets with 200K context windows, custom instructions, and artifact generation for code/document collaboration.
May 20244 releases
Golden Gate Claude Research Demo
Golden Gate Claude was available for a 24-hour research demo on May 23, 2024, showcasing interpretability research by modifying model activations to focus on the Golden Gate Bridge concept.
Golden Gate Claude
A research demo of Claude 3 Sonnet with modified features to focus on the Golden Gate Bridge, available for 24 hours. The model was adjusted to activate specific neurons related to the Golden Gate Bridge, altering its responses. This is a research demonstration only.
Mapping the Mind of a Large Language Model
Announces a research paper detailing advancements in understanding AI model interpretability, specifically mapping concepts within Claude Sonnet.
ResearchSource: Anthropic
Claude is now available in Europe
Claude is now available in Europe with web, iOS, and Team plan access, featuring Claude 3 models and multilingual support.
March 20241 release
Claude 3 Model Family Launch
Announced the Claude 3 model family including Haiku, Sonnet, and Opus with improved intelligence, speed, and cost efficiency for various applications.
November 20231 release
Introducing Claude 2.1
Introducing Claude 2.1 with 200K token context window, 2x reduction in hallucination rates, and new tool use capabilities. Enhancements include improved accuracy for long documents, system prompts, and updated pricing.
September 20231 release
Introducing Claude Pro
Announced a paid plan for Claude.ai with 5x more usage of Claude 2, file upload capabilities, and priority access during high-traffic periods.
July 20232 releases
Claude 2 Launch
Announces Claude 2, a new model with improved performance, longer responses, API access, and a public beta website.
Claude 2
Announces Claude 2 with improved performance, longer responses, API access, and beta website. Includes Bar exam/GRE scores, 100K token context window, coding improvements, and safety enhancements.
May 20231 release
Zoom Partnership and Integration of Claude
Announced partnership with Zoom to integrate Claude, Anthropic's AI assistant, into Zoom Contact Center portfolio for improved customer-facing AI products. Zoom Ventures also invested in Anthropic.
December 20221 release
Constitutional AI: Harmlessness from AI Feedback
Introduces Constitutional AI, a method for training harmless AI assistants through self-improvement without human labels, using supervised and reinforcement learning phases.
ResearchSource: Anthropic
Shipping cadence
Releases per month over the last 12 months. Busiest month: September 2026, 59 releases.
What Anthropic ships
Releases in the last 12 months by category and product.
By category
- Developer Tools51
- API39
- Product34
- Infrastructure17
- Model17
- Enterprise & Security16
By product
- Claude Developer Platform48
- Claude39
- Claude Platform17
- Claude API14
- API11
- Claude Managed Agents5
Anthropic updates FAQ
When did Anthropic last ship an update?
The most recent Anthropic update we track is “Dynamic Workflows for Claude Managed Agents”, released on October 9, 2026.
What did Anthropic release in October 2026?
Anthropic shipped 30 updates in October 2026, including “Dynamic Workflows for Claude Managed Agents”, “Compliance API Enhancements”, “Compliance API chat endpoints now return chats from the unified Claude experience” and 27 mores.
How often does Anthropic release updates?
We tracked 181 updates from Anthropic in the last 12 months, about 15 a month. The busiest month was September 2026, with 59 releases.
Where is Anthropic's official changelog?
Anthropic publishes release notes on Claude Platform Release Notes and Claude Help Center Release Notes, linked at the top of this page. Every entry here links back to the official page it came from.
Related companies
Newsletter
Stay ahead of the month's AI releases without living in every company feed.
This viewscreen turns the calendar into a recurring digest: the sharpest launches, model drops, and product moves delivered in one clear read.
Monthly signal
The standout launches, model drops, and platform moves worth tracking.
Zero feed fatigue
One tight read instead of bouncing between company blogs and timelines.
Built for operators
Clear summaries you can scan fast and forward to the team when needed.
Monthly dispatch
Subscribe to the newsletter
Get a clean monthly dispatch covering the launches that mattered, not every minor product breadcrumb.
Monthly digestMajor launchesFast summaries
One email each month, tuned for signal over noise.