llm-anthropic 0.25

Simon Willison's Blog Tools
llm-tool anthropic claude plugin python open-source

Summary

llm-anthropic 0.25 releases with support for Claude Opus 4.7, new thinking_effort and thinking_display options, increased default max_tokens, and removal of obsolete beta headers.

No content available
Original Article
View Cached Full Text

Cached at: 04/20/26, 08:26 AM

# Release: llm-anthropic 0.25 Source: https://simonwillison.net/2026/Apr/16/llm-anthropic/ Releasellm\-anthropic 0\.25 (https://github.com/simonw/llm-anthropic/releases/tag/0.25)— LLM access to models by Anthropic, including the Claude series > - New model:`claude\-opus\-4\.7`, which supports`thinking\_effort`:`xhigh`\. \#66 - New`thinking\_display`and`thinking\_adaptive`boolean options\.`thinking\_display`summarized output is currently only available in JSON output or JSON logs\. - Increased default`max\_tokens`to the maximum allowed for each model\. - No longer uses obsolete`structured\-outputs\-2025\-11\-13`beta header for older models\.

Similar Articles

llm-anthropic 0.25.1

Simon Willison's Blog

llm-anthropic 0.25.1 adds support for Claude Opus 4.8, a fast mode option, and updates default max_tokens behavior.

llm-anthropic 0.26

Simon Willison's Blog

llm-anthropic 0.26 adds support for Claude 5 models, new server-side tools (WebSearch, WebFetch, CodeExecution, AnthropicMCP), and upgrades to LLM 0.32 with streaming typed events and simplified thinking controls.

llm 0.32a3

Simon Willison's Blog

Release of llm 0.32a3, an alpha version of Simon Willison's command-line tool for interacting with large language models.

llm 0.32rc2

Simon Willison's Blog

llm 0.32rc2 fixes a dependency issue, makes GPT-5.6 Luna the default model, and adds a new 'llm openai endpoint' command for running prompts against OpenAI-compatible endpoints without configuration.

llm-echo 0.5a0

Simon Willison's Blog

llm-echo 0.5a0 release adds a -o thinking 1 option to help test against LLM 0.32a0 and higher. This plugin provides a fake echo model for automated testing.