Fast mode for Claude Opus 4.8 deprecated with removal on November 30, 2026
Developers using fast mode with claude-opus-4-8 must migrate to Claude Opus 5.5 before November 30, 2026, or their requests will return errors.
Evidence
+We've deprecated fast mode for Claude Opus 4.8, with removal on November 30, 2026. After removal, requests to claude-opus-4-8 with speed: "fast" will return an error. To continue using fast mode, migrate to Claude Opus 5.5.
Usage and Cost Analytics API launched in beta with new endpoints
Claude Console and Claude Enterprise organizations can now access token usage and cost data via new /v1/organizations/analytics/usage and /v1/organizations/analytics/cost endpoints with advanced filtering and grouping options.
Evidence
+We've launched the Usage and Cost Analytics API in beta with the analytics-usage-cost-2026-08-24 beta header. Its /v1/organizations/analytics/usage and /v1/organizations/analytics/cost endpoints report token usage and cost for Claude Console and Claude Enterprise organizations.
Maximum request size increased from 32 MB to 128 MB for Messages API and Token Counting API
Developers can now send larger requests to the Messages API and Token Counting API (up to 128 MB), enabling processing of larger documents and payloads.
Evidence
+We've raised the maximum request size for the Messages API and the Token Counting API from 32 MB to 128 MB on the first-party Claude API.
User profiles feature launched for multi-customer platforms.
Platform operators and API resellers can now associate API requests with specific end customers, enabling granular safety enforcement actions. New /v1/user_profiles endpoints available in SDKs and CLI.
Evidence
+We've launched user profiles, 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 ant CLI support the new /v1/user_profiles endpoints.
New Models API capability for thinking between tools
API users can now query whether models support thinking: {type: "between_tools"} via GET /v1/models endpoints, enabling better control over when models perform reasoning.
Evidence
+We've added capabilities.thinking.types.between_tools to the Models API. GET /v1/models and GET /v1/models/{model_id} now report whether each model accepts thinking: {type: "between_tools"}, which turns off up-front thinking.
OpenTelemetry export launched in beta for tracing Messages API and Claude Managed Agents
Developers can now integrate Claude API requests with their observability backends to trace and monitor Messages API requests and Claude Managed Agents sessions across Python, TypeScript, C#, Go, Java, PHP, and Ruby SDKs.
Evidence
+We've launched OpenTelemetry export 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.
Claude Haiku 4.5 code breaks on Haiku 5.5; manual extended thinking (budget_tokens) returns 400 error
Developers using Claude Haiku 4.5 must migrate code to Haiku 5.5, as the new version enables adaptive thinking by default and changes token counting, breaking existing implementations.
Evidence
+Code written for Claude Haiku 4.5 can break on Claude Haiku 5.5. Manual extended thinking (budget_tokens) returns a 400 error, and adaptive thinking is on by default.
Models API now includes a 'line' field to group models by their family
Developers can now use the 'line' field returned by GET /v1/models and GET /v1/models/{model_id} to group and categorize models (e.g., opus, sonnet) without parsing model IDs manually.
Evidence
+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.
Breaking changes introduced in Claude Sonnet 5.5 affecting code compatibility
Code written for Claude Sonnet 5 may break on Claude Sonnet 5.5; developers need to migrate. Changes include: thinking behavior modifications, forced tool use returning errors, computer use tool deprecation, and advisor tool restrictions
Evidence
+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"} instead of "disabled", at high effort or below. Forced tool use (tool_choice types any and tool) returns a 400 error. Thinking blocks are tied to the model and the conversation. On the Claude API and Google Cloud, the earlier computer_20251124 computer use tool isn't accepted. The advisor tool rejects Claude Opus 4.8, Claude Opus 4.7, and Claude Sonnet 5 as advisors.
Compliance API local session endpoints exit beta for Claude for Microsoft 365 in Office applications.
The Compliance API local session endpoints for Claude in Excel, PowerPoint, Word, and Outlook (office_agents) are now generally available, no longer in beta status.
Evidence
+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). See Sessions on users' machines.
Claude Fable 5 positioning changed from 'widely released model' to 'model open to all customers'.
Claude Fable 5 availability positioning has been adjusted to emphasize it is available to all customers, indicating a change in access scope or marketing messaging.
Evidence
−We've launched Claude Fable 5 (claude-fable-5), our most capable widely released model
+We've launched Claude Fable 5 (claude-fable-5), our most capable model open to all customers
Claude Opus 5.5 requires API changes: thinking field disabled, uses effort parameter, and new computer toolset
Developers must update code to remove thinking field configuration, use effort parameter for thinking depth control, and migrate to computer_toolset_20260801 on Claude API/Google Cloud (or keep computer_20251124 on Bedrock)
Evidence
+On Claude Opus 5.5, thinking can't be disabled: thinking: {"type": "disabled"} and thinking: {"type": "enabled", ...} return a 400 error. Omit the thinking field and control thinking depth with the effort parameter. tool_choice types any and tool also return a 400 error, as on Claude Fable 5.1; use auto with strict tool use. On the Claude API and Google Cloud, computer use on this model requires the computer_toolset_20260801 toolset and the earlier computer_20251124 tool returns a 400 error
Compliance API now returns transcripts from Claude in Chrome sessions in beta
Claude Enterprise organizations can now access transcripts of Claude in Chrome sessions through the existing Compliance API local session endpoints using their current Compliance Access Key and read:compliance_user_data scope.
Evidence
+The 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, with your existing Compliance Access Key and the read:compliance_user_data scope.
Claude Managed Agents permission policies now support 'auto' evaluation mode
Developers can now configure permission policies to automatically evaluate, deny, or request approval for agent and MCP tool calls, with detailed evaluation reporting through new fields in agent.tool_use and agent.mcp_tool_use events.
Evidence
+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 evaluated_permission.
Audio files now included in C2PA Content Credentials support for code execution outputs
Developers using Claude's code execution tool can now retrieve audio files with C2PA Content Credentials through the Files API, expanding the types of media files that carry content authenticity markers.
Evidence
−supported image and video files that Claude produces through the code execution tool carry C2PA Content Credentials
+supported image, video, and audio files that Claude produces through the code execution tool carry C2PA Content Credentials
Claude Enterprise Admin API, Analytics API, and Compliance API now require anthropic-version header on all requests
Developers using Claude Enterprise Admin API, Analytics API, or Compliance API must now include the anthropic-version header in every request to these endpoints, matching the requirement for the rest of the Claude API.
+The guides for the Claude Enterprise endpoints of the Admin API (user management and spend limits), the Claude Enterprise Analytics API, and the Compliance API now show the anthropic-version header; send it on every request to these endpoints, as in the rest of the Claude API. See API versions.
Ask Docs
Admin API now available in CLI and multiple SDK languages
Organization administrators can now manage organization settings, members, workspaces, API keys, rate limits, and other admin functions through ant CLI and Python, TypeScript, C#, Go, Java, PHP, and Ruby SDKs instead of curl-only access.
Evidence
+The Admin API is now available in the ant CLI and the Python, TypeScript, C#, Go, Java, PHP, and Ruby SDKs under client.beta.organization. They cover organization info, members, invites, workspaces and workspace members, API keys, rate limits, service accounts, workload identity federation issuers and rules, and customer-managed encryption keys.
Admin API user-management endpoints for Claude Enterprise are now generally available
Organizations using Claude Enterprise can now manage users, invites, groups, and custom roles via the Admin API without beta headers.
Evidence
+The Admin API user-management endpoints for Claude Enterprise (claude.ai) organizations (members, invites, groups, and custom roles) are now generally available. The anthropic-beta: ce-user-management-2026-07-13 header is no longer required on group and custom-role requests
Clarity added: Claude Haiku 3.5 retirement is API-specific, not universal
Developers using Claude Haiku 3.5 on Amazon Bedrock or Google Cloud can continue using it; only Claude API users need to migrate to Claude Haiku 4.5.
Evidence
−We've retired the Claude Sonnet 3.7 model (claude-3-7-sonnet-20250219) and the Claude Haiku 3.5 model (claude-3-5-haiku-20241022). All requests to these models will now return an error.
+We've retired the Claude Sonnet 3.7 model (claude-3-7-sonnet-20250219) and the Claude Haiku 3.5 model (claude-3-5-haiku-20241022). All requests to Claude Sonnet 3.7 will now return an error. Requests to Claude Haiku 3.5 on the Claude API will now return an error; it remains available on Amazon Bedrock and Google Cloud.
New anthropic-workspace-id response header added to Claude API
Developers can now identify the workspace behind an API response using the new wrkspc_-prefixed header, enabling better workspace management and organization tracking.
Evidence
+We've added the anthropic-workspace-id response header to the Claude API. It carries the wrkspc_-prefixed ID of the workspace that the request's API key or access token resolved to, including your organization's Default Workspace. See Identify the workspace behind an API response.
Compliance API expanded to support local sessions from Cowork and Claude Code
Claude Enterprise organizations can now retrieve transcripts and metadata from Cowork and Claude Code sessions running locally on users' machines using new endpoints: GET /v1/compliance/apps/sessions/local, GET /v1/compliance/apps/sessions/local/{session_id}, and GET /v1/compliance/apps/sessions/local/{session_id}/messages.
Evidence
+The Compliance API now returns transcripts of Cowork and Claude Code sessions that run on your users' machines, in beta for Claude Enterprise organizations. GET /v1/compliance/apps/sessions/local lists sessions across your organization, GET /v1/compliance/apps/sessions/local/{session_id} retrieves one session's metadata, and GET /v1/compliance/apps/sessions/local/{session_id}/messages returns its transcript
Removal of introductory pricing notation and terms for Claude API
Developers will no longer see the explicit introductory pricing rates ($2/$10) and the sunset date (August 31, 2026), or the standard pricing thereafter ($3/$15). This could affect pricing decisions and understanding of rate changes.
Evidence
−Input*
Output*
Write*
Read*
*Introductory pricing of $2/$10 per million input/output tokens through August 31, 2026; $3/$15 standard pricing thereafter.
Claude Opus 4.1 (claude-opus-4-1-20250805) model retired
Users and applications using Claude Opus 4.1 will receive errors and must migrate to Claude Opus 5. Researchers can apply for continued access through the External Researcher Access Program.
Evidence
+We've retired the Claude Opus 4.1 model (claude-opus-4-1-20250805). All requests to this model will now return an error. We recommend upgrading to Claude Opus 5. Researchers can request ongoing access through the External Researcher Access Program.
Claude Managed Agents update semantics now make version field optional
Users can now update agents without providing a version (unconditional update) or supply it for optimistic concurrency control with 409 error on mismatch.
Evidence
+The version field is now optional when updating a Claude Managed Agents agent. Supply it for optimistic concurrency (a mismatch returns a 409 error), or omit it to apply the update unconditionally. See Update semantics.
Users will lose access to the legacy Workbench (platform.claude.com/workbench) in the Claude Console. Saved prompts, variables, and evals from the old Workbench are not supported in the updated version. Users need to export any data they want to keep before August 17, 2026.
Evidence
+The legacy Workbench (platform.claude.com/workbench) in the Claude Console is being sunset with access ending on August 17, 2026. Saved prompts, variables, and evals are not supported in the updated Workbench. You can export any data you want to keep from the banner and under your Organizational Settings.
Mid-conversation system messages feature now generally available across multiple Claude models and platforms
Developers using Claude Fable 5, Claude Mythos 5, and Claude Opus 4.8 can now use mid-conversation system messages without beta headers on Claude API, Amazon Bedrock, and Google Cloud.
Evidence
+Mid-conversation system messages are available on Claude Fable 5, Claude Mythos 5, and Claude Opus 4.8, on the Claude API, Claude in Amazon Bedrock, and Google Cloud. No beta header is required.