Claude Platform

Expanded Access Transparency documentation for cmek_preserve events with a filter example, example event payload, and two preservation reason codes. Also clarifies that a preservation event is written whether initiated by a human reviewer or an automated safety pipeline.

You can now set an expiration when creating API keys or Admin API keys in the Claude Console, choosing a preset, custom duration, or Never. Keys with a lifetime of at least 7 days trigger an email reminder before expiration.

Added the agent-memory-2026-07-22 beta header, which changes listing memories to return results in a stable server-defined order, ignores order_by and order, restricts depth to 0 or 1, and tightens path_prefix matching. Page cursors from the old header are invalid; restart from the first page. This header replaces managed-agents-2026-04-01 on memory store endpoints, and both headers together return a 400 error.

We've restored access to Claude Fable 5 and Claude Mythos 5. See our statement for more information.

We've launched Claude Sonnet 5 (claude-sonnet-5), the next generation of our Sonnet model family, at introductory pricing of $2 / $10 per MTok through August 31, 2026 (standard $3 / $15 thereafter). Claude Sonnet 5 supports a 1M token context window, 128k max output…

Claude Managed Agents session event streams now support event deltas. Opt in with the event_deltas[] query parameter on GET /v1/sessions/{session_id}/events/stream. Listing sessions now supports backward pagination with prev_page cursor. When creating a session, you can…

We've removed fast mode for Claude Opus 4.6. Requests to claude-opus-4-6 with speed: "fast" no longer run at fast speed or premium pricing: they run at standard speed, are billed at standard rates, and do not return an error. The response's usage.speed field reports the…

Rate limits raised across the Claude API. Claude Sonnet and Claude Haiku rate limits now match Claude Opus at every usage tier. Usage tiers consolidated into three: Start, Build, and Scale. Most organizations move to a higher tier, no organization receives lower limits than…

Fast mode for Claude Opus 4.7 deprecated, with removal on July 24, 2026. After removal, requests to claude-opus-4-7 with speed: "fast" will return an error. Migrate to fast mode for Claude Opus 4.8.

MCP tunnels (research preview): the management API moved from /v1/organizations/tunnels on the Admin API to /v1/tunnels on the Claude API. The new surface uses the anthropic-beta: mcp-tunnels-2026-06-22 header and the workspace:manage_tunnels WIF scope. The previous…

The Python, TypeScript, Go, Java, Ruby, PHP, and C# SDKs now include support for code_execution_20260120, the code execution tool version that adds REPL state persistence and is the minimum version for programmatic tool calling. To adopt it, set the tool's type to…

Python, TypeScript, Go, Java, Ruby, PHP, and C# SDKs now support code_execution_20260120, the code execution tool version that adds REPL state persistence and is the minimum version for programmatic tool calling. Set tool type to code_execution_20260120; no beta header…

We've retired the Claude Sonnet 4 model (claude-sonnet-4-20250514) and the Claude Opus 4 model (claude-opus-4-20250514). All requests to these models will now return an error. We recommend upgrading to Claude Sonnet 4.6 and Claude Opus 4.8 respectively. Researchers can…

Code execution tool: Now supports code_execution_20260521, which discloses the 90-second per-cell execution time limit in the tool description so Claude can budget long-running cells. No beta header required.

The web search and web fetch tools now support version 20260318, adding a response_inclusion parameter that drops consumed result blocks from the API response for agentic workflows.

The web search tool and web fetch tool now support web_search_20260318 and web_fetch_20260318, adding a response_inclusion parameter to drop consumed result blocks from the API response for agentic workflows. No beta header is required.

The code_execution_20260521 tool version now discloses the 90-second per-cell execution time limit in the tool description, enabling Claude to budget long-running cells. No beta header is required.

• The code execution tool now supports code_execution_20260521, which discloses the 90-second per-cell execution time limit in the tool description so Claude can budget long-running cells. No beta header is required.
• The web search tool and web fetch tool now support…

The code execution tool now supports code_execution_20260521, which discloses the 90-second per-cell execution time limit in the tool description so Claude can budget long-running cells. No beta header is required.

The web search and web fetch tools now support a response_inclusion parameter that lets agentic workflows drop consumed result blocks from the API response. No beta header is required for the new tool versions.