What changed in the Anthropic API

If you operate a platform or gateway, or resell API access, a safety enforcement action can then apply to one end customer instead of your whole organization.

 are out of beta for Claude Enterprise organizations.
, so you can tell Anthropic which of your end customers each API request is for. If you operate a platform or gateway, or resell API access, a safety enforcement action can then apply to one end customer instead of your whole organization. The SDK and the 
/v1/organizations/webhooks
 register webhook endpoints for two event types: one for when Anthropic restricts or bans one of your organization's user profiles, and one for when the restriction or ban ends. To request access, contact your Anthropic account team.

You can export any data you want to keep from the banner

capabilities.thinking.types.between_tools
thinking: {type: "between_tools"}
, which turns off up-front thinking. See 
, or earlier turns can also return a 400 error. Adaptive thinking is on by default, so a response can begin with 
 in the Console. You can export any data you want to keep from the banner and under your 
In the Console, users with the Developer role can now access the 

Requests to Claude Opus 5.5 no longer require

 now returns the text each organization member saved as 
 in Cowork, in beta for Claude Enterprise organizations. Call 
GET /v1/compliance/organizations/{org_uuid}/user_preferences
 with your existing Compliance Access Key and the 
 is out of research preview for Claude Opus 5.5, and you no longer need to request access. Requests to Claude Opus 5.5 no longer require the 
 beta header, and requests that still send it keep working. Claude Opus 5.5 now has a default fast mode rate limit in the Start, Build, and Scale usage tiers. Fast mode for Claude Opus 5 and Claude Opus 4.8 remains in research preview. Keep sending the beta header with fast mode requests to those models.
 for Claude Opus 5.5 is now available on 
 for deployments hosted on Azure.

The Claude SDK emits spans in Python, TypeScript, C#, Go, Java, PHP, and Ruby.

 in beta, so you can trace Messages API requests and Claude Managed Agents sessions in your observability backend. The Claude SDK emits spans in Python, TypeScript, C#, Go, Java, PHP, and Ruby. To get spans from the API, 
add an OpenTelemetry destination

is not supported, and manual extended thinking budgets and assistant prefill are not supported (both return a 400 error).

You can now let a Claude Managed Agents agent use 
 beta header. For work with many pieces, such as reviewing hundreds of documents, the agent can write a workflow. A workflow is a program that runs many agents in phases and combines their results. The server runs it in the background as a workflow run. To 
turn dynamic workflows on
{"type": "multiagent_20261001", "workflows": {"type": "enabled"}}
. To guide the agent, tell it in its system prompt when to start a run. Follow each run through 
 events on the session's event stream. See 
The Compliance API can now download Claude Docs documents as Word files, in beta. See 
Retrieve files and artifacts
), the sampling parameters 
, and assistant message prefill can each return a 400 error. 
 also returns a 400 error: use the 
 toolset on the Claude API and Google Cloud, or 
 on Amazon Bedrock. A request that sends thinking blocks back after a change to 
, or earlier turns can also return a 400 error. On Amazon Bedrock, structured outputs aren't available for Claude Haiku 5.5. Adaptive thinking is on by default, so a response can begin with 
 blocks, and their text is omitted unless you set 
. The same text also counts as more tokens. 

(research preview) now supports Claude Opus 5.5, Claude Fable 5.1, and Claude Sonnet 5.5.

 chat endpoints now also return chats from the unified Claude experience, in beta for Claude Enterprise organizations, with your existing Compliance Access Key. See 
Retrieve and delete chats, files, and projects
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. See 
), our most capable model tuned for high-volume and latency-sensitive work. It has a 
, 128k max output tokens, and 
. It's available on the Claude API, 
What's new in Claude Haiku 5.5
Code written for Claude Haiku 4.5 can break on Claude Haiku 5.5. Manual extended thinking (
) returns a 400 error, and adaptive thinking is on by default, so a response can begin with 
 blocks. The same text also counts as more tokens. See the 
. For model-specific prompting patterns, see 
Prompting Claude Haiku 5.5
The Python and TypeScript SDKs now include classes, in beta, for the 
. You subclass one and write one method per tool against your own browser or desktop automation. The SDK runs the tool loop, the URL and file policies you set for the browser, and your approval callbac

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

capabilities.thinking.types.disabled
 now report whether each model accepts 
thinking: {type: "disabled"}
, which turns thinking off. See 
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 
 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: call them under 
. These releases also keep 

now return the model line each model belongs to.

 now return the model line each model belongs to. Claude Opus 4.5 and Claude Opus 4.6 both report 
 to group models without parsing their IDs. 
 for a model that belongs to no line. See 

We announced the deprecation of the Claude Sonnet 4.5 model

We announced the deprecation of the Claude Sonnet 4.5 model (
), with retirement on the Claude API scheduled for November 30, 2026. We recommend migrating to 
) returns a 400 error. Thinking blocks are tied to the model and the conversation. On the Claude API and Google Cloud, the earlier 

Thinking blocks that Claude Sonnet 5.5 produces work only in the account that produced them, or in an account linked to it.

). It's available on the Claude API, 
. For its context window, output limits, and prices, see the 
Claude Sonnet 5.5 model page
Code written for Claude Sonnet 5 can break on Claude Sonnet 5.5 in five ways. To turn off up-front thinking, send 
thinking: {"type": "between_tools"}
 effort or below. Forced tool use (
) returns a 400 error. Thinking blocks are tied to the model and the conversation. 
On the Claude API and Google Cloud, the
 computer use tool isn't accepted. The advisor tool rejects Claude Opus 4.8, Claude Opus 4.7, and Claude Sonnet 5 as advisors. See 
What's new in Claude Sonnet 5.5
 for before-and-after requests. For model-specific prompting patterns, see 
Prompting Claude Sonnet 5.5
Thinking blocks that Claude Sonnet 5.5 produces work only in the account that produced them, or in an account linked to it. When another account sends one of these blocks, the API drops the block before the model sees it, and the request succeeds. Blocks from earlier models aren't affected. See 

no longer returns file names, project document names, or artifact titles.

We're resuming billing for refusals that arrive before any output when 
, the categories where we measure low volumes of false positives. Mid-stream refusals were already billed. Refusals billed under this change are charged like any other request, at the rates of the model that ran it. Refusals before any output in other categories are still not billed, and fallback credit is unchanged. This change applies on all platforms. See 
 local session endpoints are out of beta for Claude for Microsoft 365 sessions in Excel, PowerPoint, Word, and Outlook (
 no longer returns file names, project document names, or artifact titles. The 
 fields on file, project document, and artifact activities are now always empty or omitted, including on activities recorded before this change. To look up a name or title by the ID on the activity, use a Compliance Access Key with the 
Understand the Activity object
 is out of beta on the Claude API and no longer requires the 
 object on a Messages request to opt in; requests that still send the header work as before. Responses from 
 when the request did not include the 
, a response to a request that sends the 
 beta header now always includes the 
 when t

The same header covers adding and removing tools by reference.

), a model for long-running agentic coding and knowledge work. It has a 
, at $4 / $20 USD per MTok (Claude Opus 5 is $5 / $25). Claude Opus 5.5 is available on the Claude API, 
What's new in Claude Opus 5.5
On Claude Opus 5.5, thinking can't be disabled: 
thinking: {"type": "enabled", ...}
 return a 400 error. Omit the 
 field and control thinking depth with the 
 also return a 400 error, as on Claude Fable 5.1; use 
. On the Claude API and Google Cloud, 
 on this model requires the 
 tool returns a 400 error; on Amazon Bedrock, 
 (research preview) is available for Claude Opus 5.5 on the Claude API.
Tools can now be defined inside a 
mid-conversation system message
, in beta on the Claude API with the 
 block can carry the tool's full definition (
tool: {"type": "tool_definition", "definition": {...}}
), so you can add a tool, change its schema, or move a server tool to a newer version without editing 
 or invalidating the prompt cache. The same header covers adding and removing tools by reference.
 beta header as well, the definition can be an MCP toolset, and a response records each server's fetched tool list in an 
 block, which pins that list when you send it back.

It names a thinking block that failed the prefix check on a request where the API doesn't enforce that check: on Claude Fable 5.1, for example, a request from an account created before August 31, 2026

 response field gains a second entry type, 
thinking_mismatch_allowed
. It names a thinking block that failed the prefix check on a request where the API doesn't enforce that check: on Claude Fable 5.1, for example, a request from an account created before August 31, 2026, with 
 unset. The block still reaches the model unchanged. Log these entries to find history edits in production traffic before you opt into enforcement. See 
Set the mismatch behavior and read 

local session endpoints now also return transcripts of Claude in Chrome sessions

 local session endpoints now also return transcripts of Claude in Chrome sessions (

On later requests, send that block first, in place of those messages. You choose when to compact, the request can run in the background, and you can keep recent turns word for word after the summary…

compact a conversation on demand
 on the Claude API, in beta with the 
 beta header. Send the top-level 
 parameter, and the API returns a signed 
 block that summarizes the messages you sent. On later requests, send that block first, in place of those messages. You choose when to compact, the request can run in the background, and you can keep recent turns word for word after the summary. On models with preserved thinking, the thinking in those kept turns can stay valid.

You can follow the session live, send messages, and allow or deny tool calls that are waiting for approval.

Claude Managed Agents permission policies now include 
: the server evaluates each agent or MCP tool call and runs it, denies it, or pauses for your approval. 
 events report how each call was evaluated in an 
Let the server evaluate each call with 
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 
 to serve the Claude Console's session viewer locally and open the session there instead. See 
Connect to a Managed Agents session from your terminal

for Claude Fable 5.1, Claude Mythos 5.1, and Claude Opus 5, with the same

 changes, in beta, are also available on 
 for Claude Fable 5.1, Claude Mythos 5.1, and Claude Opus 5, with the same