Anthropic new features in September 2026

Everything Anthropic shipped in September 2026: 36 releases, grouped by product, each linked to Anthropic's release notes.

August 2026

GA 4Preview 1Beta 11Pricing 1Update 19

Claude Platform21

Claude PlatformSep 22, 2026

Claude API beta lets tools be defined inside mid-conversation system messages

With the inline-tools-2026-09-15 beta header, a tool_addition block can carry a full tool definition, so agents can add tools, change schemas or upgrade server tools mid-conversation without invalidating the prompt cache. The same header covers adding and removing tools by reference.

BetaLLMsMCPData integrationSource

Claude PlatformSep 22, 2026

Claude Opus 5.5 breaking changes: thinking can't be disabled, forced tool_choice removed

On Claude Opus 5.5, disabling or manually enabling thinking returns a 400; control depth with the effort parameter instead. tool_choice any and tool also return 400, so use auto with strict tool use, and computer use requires the computer_toolset_20260801 toolset.

UpdateLLMsDeveloper toolsSource

Claude PlatformSep 22, 2026

Claude Opus 5.5 launches with a 1M token context window and 128k output

Anthropic released Claude Opus 5.5 for long-running agentic coding and knowledge work, with a default 1M token context, 128k max output and always-on adaptive thinking. It costs $4 / $20 per MTok, below Claude Opus 5, and is on the Claude API, Bedrock, Google Cloud and Microsoft Foundry.

GAAI agentsLLMsDeveloper toolsSource

Claude PlatformSep 22, 2026

Fast mode research preview available for Claude Opus 5.5 on the Claude API

Fast mode, in research preview, can now be used with Claude Opus 5.5 on the Claude API.

PreviewLLMsDeveloper toolsSource

Claude PlatformSep 18, 2026

Compliance API adds Claude in Chrome session transcripts in beta

Compliance API local session endpoints now return transcripts of Claude in Chrome sessions (product_surface claude_in_chrome) for Claude Enterprise organizations, using the existing Compliance Access Key and read:compliance_user_data scope.

BetaLLMsGovernanceDeveloper toolsSource

Claude PlatformSep 14, 2026

Claude API input_transformations adds thinking_mismatch_allowed entries in beta

Under the thinking-binding-controls-2026-08-01 beta header, input_transformations can report thinking_mismatch_allowed for thinking blocks that failed an unenforced prefix check. Logging these entries helps find conversation history edits in production traffic.

BetaLLMsObservabilityDeveloper toolsSource

Claude PlatformSep 14, 2026

Messages API adds on-demand conversation compaction in beta

With the compact-2026-09-04 beta header, the top-level compaction parameter returns a signed compaction block summarizing prior messages, which you send in their place on later requests. You choose when to compact, it can run in the background, and recent turns can be kept verbatim.

BetaLLMsDeveloper toolsSource

Claude PlatformSep 10, 2026

Claude Managed Agents permission policies add an auto mode

A new auto permission policy lets the server evaluate each agent or MCP tool call and run it, deny it, or pause for approval. agent.tool_use and agent.mcp_tool_use events now report how each call was evaluated.

UpdateAI agentsLLMsMCPSource

Claude PlatformSep 10, 2026

ant CLI 1.32.0 adds sessions connect for live Managed Agents sessions

ant beta:sessions connect attaches a terminal to a Claude Managed Agents session so you can follow it live, send messages and approve or deny pending tool calls. --web serves the Console session viewer locally.

BetaAI agentsLLMsDeveloper toolsSource

Claude PlatformSep 3, 2026

Per-message effort changes beta expands to Claude on Google Cloud

Per-message effort changes are now available in beta on Google Cloud for Claude Fable 5.1, Claude Mythos 5.1 and Claude Opus 5, using the same mid-conversation-output-config-2026-07-01 beta header.

BetaLLMsSource

Claude PlatformSep 3, 2026

ant CLI 1.30.0 adds ant apply to manage agents and resources as code

ant apply creates and updates agents, environments, skills, memory stores and deployments from files in a repository after you approve a printed plan. A committed claude-lock.json lets later local or CI runs update the same resources instead of creating new ones.

UpdateAI agentsLLMsPricingSource

Claude PlatformSep 1, 2026

Claude API adds turn-scoped system messages in beta

Setting clear_at next_user_message on a mid-conversation system message renders it for the current turn only, then keeps it in history at no token cost. Per-turn reminders no longer accumulate or invalidate the prompt cache and later thinking blocks.

BetaPricingSource

Claude PlatformSep 1, 2026

Claude API thinking.display adds an updates mode in beta

The new updates value, behind the thinking-display-updates-2026-08-18 header, returns reasoning empty like omitted but surfaces the short progress updates Fable and Mythos models write between tool calls as text.

BetaLLMsSource

Claude PlatformSep 1, 2026

Claude Enterprise Admin, Analytics and Compliance API guides add anthropic-version

Guides for the Claude Enterprise Admin API endpoints, the Enterprise Analytics API and the Compliance API now show the anthropic-version header, which should be sent on every request as with the rest of the Claude API.

UpdateLLMsGovernanceDeveloper toolsSource

Claude PlatformSep 1, 2026

Claude Fable 5.1 and Mythos 5.1 reject tool_choice any and tool

On Claude Fable 5.1 and Claude Mythos 5.1, tool_choice types any and tool return a 400 error while auto and none are unchanged. Use strict tool use or structured outputs to guarantee schema-conformant tool inputs.

UpdateLLMsSource

Claude PlatformSep 1, 2026

Claude Fable 5.1 and Mythos 5.1 require 30-day data retention

Like Claude Fable 5, both new models require 30-day data retention and are not available under zero data retention unless Anthropic expressly authorizes it, which affects teams with ZDR agreements.

UpdateLLMsSource

Claude PlatformSep 1, 2026

Claude Fable 5.1 launches alongside Claude Mythos 5.1 with 1M token context

Claude Fable 5.1 succeeds Claude Fable 5 for long-running agentic coding, knowledge work and research, with Claude Mythos 5.1 for Project Glasswing participants. Both have a 1M context, 128k output and adaptive thinking at $10 / $50 per MTok, with cache reads cut to $0.25 per MTok.

GAAI agentsLLMsDeveloper toolsSource

Claude PlatformSep 1, 2026

Claude Fable 5.1 output carries text watermarks and C2PA Content Credentials

Text from Claude Fable 5.1 and Mythos 5.1 carries Anthropic's text watermark, and image, video and audio files produced via code execution carry C2PA Content Credentials when retrieved through the Files API. No request changes are needed.

UpdateLLMsDeveloper toolsSource

Claude PlatformSep 1, 2026

Claude Fable 5.1 thinking blocks only replay to the same or newer models

Thinking blocks from Claude Fable 5.1 and Mythos 5.1 can't be read by earlier models and are dropped if replayed to them. Fable 5.1 also checks that nothing before a replayed block has changed, which matters for agents that edit conversation history.

UpdateLLMsDeveloper toolsSource

Claude PlatformSep 1, 2026

Per-message effort changes enter beta on the Claude API

A role system message carrying output_config.effort inside messages changes effort for later turns while preserving the prompt cache, on Claude Fable 5.1, Mythos 5.1 and Opus 5. Requests need the mid-conversation-output-config-2026-07-01 beta header.

BetaLLMsDeveloper toolsSource

Claude PlatformSep 1, 2026

Prompt cache reads on Claude Fable 5.1 and Mythos 5.1 drop to $0.25 per MTok

Cache reads on Claude Fable 5.1 and Claude Mythos 5.1 cost 0.025x the base input price instead of the usual 0.1x, which lowers the cost of cache-heavy agent workloads. Cache write pricing is unchanged.

PricingLLMsPricingSource

Claude Code11

Claude CodeSep 23, 2026

Claude Code 2.1.281

What's changed Added Claude apps gateway support for newer Claude Desktop keys in desktop policy blocks, including blockReadsOutsideWorkingDirectories and disableBypassPermissionsMode Added assume_role on Claude apps gateway Bedrock upstreams: the gateway calls Bedrock as an IAM role it assumes through STS, in another AWS account if needed, optionally one...

UpdateUnverifiedLLMsMCPStreamingSource

Claude CodeSep 22, 2026

Claude Code 2.1.280

What's changed Added Claude Opus 5.5 ( claude-opus-5-5 ), now the default Opus model , 1M context, $4/$20 per Mtok with $0.20/Mtok cache reads Added mouse support to more lists in fullscreen mode: the wheel scrolls the /skills list, and a skill's state options in /plugin can be clicked Added CLAUDE_CODE_MAX_MCP_DESCRIPTION_LENGTH to change the...

UpdateLLMsMCPSecuritySource

Claude CodeSep 19, 2026

Claude Code 2.1.278

What's changed Changed auto mode for Claude API and Enterprise users, and on Bedrock, Vertex, Foundry and gateways, to default to the server-side classifier, which does not charge for classifier overhead ( CLAUDE_CODE_AUTO_MODE_SERVER=0 opts out on Bedrock, Vertex, Foundry and gateways); warns on billed fallback.

UpdateLLMsPricingDeveloper toolsSource

Claude CodeSep 18, 2026

Claude Code 2.1.276

What's changed Fixed every request failing with 400 … Input tag 'advisor_20260301' when ANTHROPIC_BASE_URL points at a proxy or gateway (2.1.275 regression)

UpdateLLMsGovernanceCoding agentsSource

Claude CodeSep 18, 2026

Claude Code 2.1.277

What's changed Added AGENTS.md support: in a project with no CLAUDE.md, Claude Code reads AGENTS.md instead; change it under "Project instructions" in /config (not yet on Bedrock, Vertex or Foundry) Added CLAUDE_GATEWAY_PROXY_IS_EGRESS_BOUNDARY=1 for Claude apps gateways whose only egress is a forward proxy: every outbound request hands the proxy the...

UpdateAI agentsLLMsStreamingSource

Claude CodeSep 17, 2026

Claude Code 2.1.274

What's changed Added a visible warning when memory usage is critical, with steps to free memory or restart safely Added CLAUDE_CODE_MCP_STARTUP_WAIT_MS to bound how long the first non-interactive turn waits for connecting MCP servers ( 0 = don't wait) Added effort attribute to the claude_code.llm_request OpenTelemetry trace span, matching the api_request...

UpdateAI agentsLLMsMCPSource

Claude CodeSep 17, 2026

Claude Code 2.1.275

What's changed Added the signed-in account to Claude apps gateway sign-in: when the gateway names it, you confirm it before the credential is saved, and /status shows it Added a send-now key (ctrl+enter, or ctrl+x ctrl+s) that interrupts the current turn and sends all queued messages at once; sent and queued messages show in gray until the model receives...

UpdateLLMsMCPStreamingSource

Claude CodeSep 15, 2026

Claude Code 2.1.272

What's changed Bug fixes and reliability improvements

UpdateLLMsCoding agentsSource

Claude CodeSep 15, 2026

Claude Code 2.1.273

What's changed Added x-claude-code-request-class , x-claude-code-agent-type , x-claude-code-prev-tool-durations , x-claude-code-compaction and x-claude-code-context-compacted request headers for LLM gateways; opt in with CLAUDE_CODE_GATEWAY_HINT_HEADERS=1 Added a notification when an MCP server disconnects mid-session and automatic reconnection gives up...

UpdateAI agentsLLMsMCPSource

Claude CodeSep 14, 2026

Claude Code 2.1.271

What's changed Added fast mode in Claude Code Remote sessions (cloud and self-hosted runners): the host's fast-mode setting or /fast typed in the session applies where your organization allows it Added mouse support to the /config panel in fullscreen mode: the wheel scrolls the settings list, a click on a setting's value changes it, and the row under the...

UpdateAI agentsLLMsGovernanceSource

Claude CodeSep 12, 2026

Claude Code 2.1.270

What's changed Fixed read-only git commands in Bash unexpectedly asking for permission after a session had been running for a while (regression in 2.1.269)

UpdateLLMsSecurityCoding agentsSource

Claude apps4

Claude appsSep 22, 2026

Claude Opus 5.5 launched, first model in the Claude 5.5 family

Anthropic launched Claude Opus 5.5, the first model in the Claude 5.5 family. It performs at the level of Claude Fable 5.1 on most work and costs 40% less to run than Opus 5.

GALLMsPricingSource

Claude appsSep 15, 2026

Salesforce in Claude plugin (beta) with 37 pre-built sales skills

Salesforce in Claude is a plugin that brings a seller's accounts, opportunities, and pipeline into Claude with 37 pre-built sales skills. Available in beta on all paid plans for organizations Salesforce approves through its beta sign-up.

BetaLLMsPricingData integrationSource

Claude appsSep 10, 2026

Smart reports (beta) on Claude Enterprise analyze team usage and costs

Smart reports analyze how a team uses Claude and report on the work done, what it costs, where sessions hit friction, and which repeated patterns are worth packaging as shared skills. Available in beta on Claude Enterprise plans.

BetaLLMsPricingSource

Claude appsSep 1, 2026

Claude Fable 5.1 and Claude Mythos 5.1 launched

Anthropic launched Claude Fable 5.1 and Claude Mythos 5.1, described as its most advanced models for coding and knowledge work.

GALLMsSource