The `llm_eval` Dart package has released version 0.3.1 to address a critical bug in its `Check.isValidJson` function. Previously, this function failed to recognize valid JSON when it was enclosed within Markdown code fences, a common output format for many LLMs. This led to spurious failures in continuous integration tests, misrepresenting correct model outputs as errors. The fix involves modifying the `Check.isValidJson` function to first attempt decoding the raw output and, if that fails due to a `FormatException`, to then extract and decode JSON specifically from within Markdown code fences. AI
IMPACT Improves reliability for developers using LLMs for structured output testing.
RANK_REASON This is a bug fix and minor version update for a niche developer tool, not a new product release or significant industry event.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
