from any2any_trainer.models.recommendation import SemanticIDRecommendationModel

# Load model
model = SemanticIDRecommendationModel.from_pretrained("zjkarina/omniRecsysLLM_semanticIDsmodality")

# Generate recommendations with semantic IDs
recommendations = model.predict_next_item(
    text="The user bought jeans and a t-shirt",
    id_ids=[12345, 67890],  # Item IDs from purchase history
    top_k=5,
    use_semantic_ids=True
)

Training

Base model: Qwen2.5-Omni-7B
Item vocabulary size: 709,036
ID embedding dimension: 512
VQ-VAE codebook size: 10,000
VQ-VAE codebook dimension: 256
Dataset: Amazon Fashion 2023 Full

The model was trained on the Amazon Fashion 2023 dataset using semantic IDs generated via VQ-VAE.