sugiv
/

cardvaultplus

Image-Text-to-Text

vision-language

card-extraction

mobile-optimized

continual-learning

structured-data

Eval Results (legacy)

Model card Files Files and versions

cardvaultplus / added_tokens.json

sugiv's picture

Upload production CardVault+ SmolVLM model - 4 epochs trained, 0.000133 validation loss

ed17412 verified 9 months ago

history blame contribute delete

92 Bytes

	{
	"<end_of_utterance>": 49154,
	"<fake_token_around_image>": 49152,
	"<image>": 49153
	}