Hub mirrors of Benjamin Kiessling's multilingual PP-OCRv6 line-recognition family for Kraken: tiny, small, and medium.
Small Models for GLAM
Team
community
AI & ML interests
None defined yet.
Recent Activity
View all activity
Fine-tuning a small open VLM (NuExtract-3, 4B) to extract structured records from archival index cards across collections.
-
small-models-for-glam/index-card-extractor-4b-v0.1
Image-Text-to-Text β’ 5B β’ Updated β’ 17 β’ 2 -
davanstrien/teklia-index-cards-kie
Viewer β’ Updated β’ 486 β’ 16 -
davanstrien/southborough-death-cards
Viewer β’ Updated β’ 299 β’ 27 -
davanstrien/rubenstein-probe-input
Viewer β’ Updated β’ 30 β’ 28
Hub mirrors of Benjamin Kiessling's multilingual PP-OCRv6 line-recognition family for Kraken: tiny, small, and medium.
Fine-tuning a small open VLM (NuExtract-3, 4B) to extract structured records from archival index cards across collections.
-
small-models-for-glam/index-card-extractor-4b-v0.1
Image-Text-to-Text β’ 5B β’ Updated β’ 17 β’ 2 -
davanstrien/teklia-index-cards-kie
Viewer β’ Updated β’ 486 β’ 16 -
davanstrien/southborough-death-cards
Viewer β’ Updated β’ 299 β’ 27 -
davanstrien/rubenstein-probe-input
Viewer β’ Updated β’ 30 β’ 28
A collection of finetunes of Qwen 3 VL. These models were finetuned on the CATMuS dataset via TRL SFT.