Last released Jul 25, 2026
Photo metadata from vision LLMs: verbatim transcription of what's written on scans, plus captions, keywords, and cautious date/location guesses.
Supported by
