Downloads · 30 days
8
6% of all-time downloads
goxq/q-geomem-llava-qwen2-vsi
q-geomem-llava-qwen2-vsi is a image-text-to-text model from goxq. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Downloads · 30 days
8
6% of all-time downloads
All-time downloads
134
Public
Repo size
1.4 GB
Likes
0
Public
Click a slice to open those files.
.bin1.4 GB · 100%
From the Hugging Face model README
This repository provides the Q-GeoMem LLaVA-Qwen2 VSI checkpoint for video spatial reasoning.
Q-GeoMem is a question-guided geometric memory model designed for spatial reasoning over videos. This checkpoint is released as a LoRA adapter plus non-LoRA trainable weights.
lmms-lab/LLaVA-Video-7B-Qwen2google/siglip-so400m-patch14-384Salesforce/blip2-itm-vit-gqgeomemK=832adapter_config.json: PEFT LoRA configurationadapter_model.bin: LoRA adapter weightsnon_lora_trainables.bin: Q-GeoMem memory/projector trainable weightsconfig.json: Q-GeoMem model configurationgeneration_config.json: generation configurationPlease follow the evaluation instructions at GitHub.