◉CLAUDE LABJP
●2.1.269 — A broad release with 98 CLI changes. The one worth your attention is claude plugin eval, which finally lets you measure whether a plugin earns its place●EVAL — Every case runs three times with the plugin loaded and three times without, so the number you get back is the difference the plugin makes, not just a pass or a fail●GRADER — Three kinds of grader are available: a regex over the reply, a check for whether a particular tool was called, and a rubric judged by a second model●DIFF — The Bash tool result now includes a diff of the files a command changed. Turn it on with bashEditDiffEnabled●STYLE — /output-style [name] lists and switches output styles, including over Remote Control and in cloud or headless sessions●OTEL — OTEL_METRICS_INCLUDE_REPOSITORY tags metrics and events with vcs. repository attributes, and commit events carry vcs.ref.head●2.1.269 — A broad release with 98 CLI changes. The one worth your attention is claude plugin eval, which finally lets you measure whether a plugin earns its place●EVAL — Every case runs three times with the plugin loaded and three times without, so the number you get back is the difference the plugin makes, not just a pass or a fail●GRADER — Three kinds of grader are available: a regex over the reply, a check for whether a particular tool was called, and a rubric judged by a second model●DIFF — The Bash tool result now includes a diff of the files a command changed. Turn it on with bashEditDiffEnabled●STYLE — /output-style [name] lists and switches output styles, including over Remote Control and in cloud or headless sessions●OTEL — OTEL_METRICS_INCLUDE_REPOSITORY tags metrics and events with vcs. repository attributes, and commit events carry vcs.ref.head
← Back to Blog

Claude Opus 4.6 Released — The New Flagship Model

Claude Opus 4.6new modelreleaseAnthropic

Claude Opus 4.6 Arrives

Anthropic has released its latest flagship model, Claude Opus 4.6, positioned as the most intelligent model for agent building and coding.

Key Features

Industry-Leading Coding Performance

Claude Opus 4.6 achieves top scores on coding benchmarks, with significant improvements in complex multi-file editing, architecture design, and debugging.

Expanded Output and Context

Maximum output has been expanded to 128K tokens (up from 32K in Opus 4). A beta 1M token context window is also available.

Extended Thinking and Adaptive Thinking

In addition to Extended Thinking, a new Adaptive Thinking mode has been introduced. It automatically adjusts reasoning depth based on problem complexity without requiring budget_tokens.

API Access

The API ID is claude-opus-4-6, available on all platforms (Claude API, AWS Bedrock, Google Vertex AI).

Pricing

Input: $5 / million tokens, Output: $25 / million tokens. While more expensive than Sonnet 4.6, it may be more cost-effective for complex tasks that require fewer interactions.

Summary

Claude Opus 4.6 represents major progress especially in agent development and coding. Try it for projects demanding the highest quality.