Table A1.

List of models tested during the preliminary phase and their respective scores: low, medium, high

ModelTypeVisual processingMetadata output
qwen2.5-VL-7BmOpen sourceMediumMedium
gemma-3-12b-it-qatOpen sourceMediumMedium
llava-v1.5-7bOpen sourceLowLow
pixtral-12b-GGUFOpen sourceHighMedium
VisionText Extractor GPT (via OpenAI)Commercial, free tier trial accessLowLow
GPT-4o (via OpenAI)Commercial, free tier trial accessHighMedium
claude-3-opus-20240229Commercial, paidMediumMedium
claude-3-5-haiku-20241022Commercial, paidMediumLow
claude-3-7-sonnet-20250219Commercial, paidHighHigh
Source(s): Authors’ own work

or Create an Account

Close Modal
Close Modal