# Multimodal Model

SpaceOm-GGUF is a multimodal model focusing on visual question answering tasks and performs excellently in spatial reasoning.

This model is a GGUF format conversion based on Qwen2-VL-7B-Captioner-Relaxed, optimized for image-to-text tasks and supports running via tools like llama.cpp and Koboldcpp.

An image captioning model based on the ViT-GPT2 architecture, capable of generating natural language descriptions for input images.

An image captioning model based on the ViT-GPT2 architecture, capable of generating natural language descriptions for input images.

A multimodal table detection model fine-tuned based on the Florence-2 model, capable of precisely locating table areas in images.

This model is a fine-tuned version of google/paligemma-3b-pt-224 on a subset of the VQAv2 dataset, specializing in visual question answering tasks.

CheXagent is a foundational model focused on chest X-ray interpretation, designed to assist in medical imaging analysis.

This is a vision encoder-decoder model based on ViT and Turkish GPT-2 for generating Turkish image captions.

CheXagent is a foundation model specifically designed for chest X-ray interpretation, capable of automatically analyzing and interpreting chest X-ray images.

ChartLlama is a multimodal model based on the LLaVA-1.5 architecture, specializing in chart understanding and analysis tasks.

This model is a fine-tuned version of Salesforce's BLIP image captioning base model on the SageMaker platform, primarily used for image caption generation tasks.

An image captioning model based on Swin Transformer and AraGPT2 architecture, capable of generating textual descriptions for input images.

A vision-language model fine-tuned on image folder datasets based on microsoft/git-base, primarily used for image caption generation tasks

This is a sharded version of the BLIP-2 model implemented with Flan T5-xl for image-to-text tasks such as image captioning and visual question answering. Sharding allows it to be loaded in low-memory environments.

An image caption generation model based on the VisionEncoderDecoder architecture, capable of converting input images into natural language descriptions.

Korean CLIP model trained via knowledge distillation, supporting Korean and English multimodal understanding

A version fine-tuned on the WildReceipt dataset based on the LayoutLMv3-base model, designed for receipt key information extraction tasks

A vision-language model based on ViT-GPT2 architecture for image-to-text tasks