taesiri/llava-videogame-qa-lora-wip

LLaVA-VideoGameVQA - Work In Progress - Model Card

Model details

Model type: LLaVA is an open-source chatbot trained by fine-tuning LLaMA/Vicuna on GPT-generated multimodal instruction-following data. It is an auto-regressive language model, based on the transformer architecture.

Model date: LLaVA-v1.5-13B-LoRA was trained in December 2023.

LoRA Weights

• Checkpoint 1 trained on 28K question-answering pairs. Base Model: liuhaotian/llava-v1.5-13b
• Checkpoint 5 trained on 74K question-answering pairs. Base Model: liuhaotian/llava-v1.5-13b
• Checkpoint 8 trained on 185K question-answering pairs. Base Model: liuhaotian/llava-v1.5-13b

How to run

python -m llava.serve.model_worker --host 0.0.0.0 --controller http://localhost:10000 --port 40000 --worker http://localhost:40000 --model-path  ./lora-checkpoints-8 --model-base liuhaotian/llava-v1.5-13b