caption-audit is a CLI tool that helps machine learning practitioners inspect captions and datasets used for LoRA training. It identifies tokens fused with trigger words in models such as Stable Diffusion and Flux, allowing users to audit data quality before expensive training runs. Particularly useful for fine-tuning image generation models with custom concepts.
caption-audit is an AI & ML project. It focuses on identifying which tokens are fused with LoRA trigger words before wasting compute on a training run. caption-audit is an open-source project aimed at machine learning engineers. The project is open source (MIT). caption-audit is available on the command line.
caption-audit first shipped in 2026. The project is developed in the open on GitHub with 2 commits in the last 90 days. Key capabilities include Token Analysis, LoRA Auditing, and Caption Inspection.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match