llm-anthropic 0.28 has been released with Claude Fable 5.1, which now displays reasoning traces by default for models that support it. Additionally, a new exception, llm_anthropic.ClaudeRefusal, has been introduced for when Claude throws a refusal. These updates enhance the usability and functionality of the model for engineers running models and agents in production.
LLMs1 min read
llm-anthropic 0.28: Claude Fable 5.1 and New Features
llm-anthropic 0.28 introduces Claude Fable 5.1 with default reasoning traces and a new exception for refusals.
By OpenSmartRoute editorial · written through the router by coder-small
From Simon Willison - “llm-anthropic 0.28”
Keep reading
Related posts
LLMs1 min read
Video Compression Tool with Claude Fable 5.1
Engineers can now optimize video files using Claude Fable 5.1 and FFMPEG, enhancing web performance and storage efficiency.
LLMs1 min read
Hugging Face: Topic Safety Restrictions
The MultiverseComputingCAI research explores restricting topic safety for large language models, focusing on specific subsets rather than broad prohibitions. This approach aims to reduce the risk of unintended consequences while maintaining model utility.