llm_vector_server/applications/config/model_config.py

MODEL_CONFIG = {
    "Qwen3-Embedding-0.6B": {"url": "http://vllm-0.6b:8000/v1/embeddings", "dim": 1536},
    "Qwen3-Embedding-4B": {"url": "http://vllm-4b:8000/v1/embeddings", "dim": 1536},
    "Qwen3-Embedding-8B": {"url": "http://vllm-8b:8000/v1/embeddings", "dim": 1536},
}

LOCAL_MODEL_CONFIG = {
    "Qwen3-Embedding-0.6B": "models/Qwen3-Embedding-0.6B",
    "Qwen3-Embedding-4B": "models/Qwen3-Embedding-4B",
    "Qwen3-Embedding-8B": "models/Qwen3-Embedding-8B",
}

DEFAULT_MODEL = "Qwen3-Embedding-0.6B"


__all__ = ["MODEL_CONFIG", "DEFAULT_MODEL", "LOCAL_MODEL_CONFIG"]