backend/app/api/cache.py

"""
Cache Statistics Endpoint

Provides real-time statistics about the model cache.
"""
from fastapi import APIRouter
from app.core.model_cache import model_cache

router = APIRouter(prefix="/cache", tags=["cache"])


@router.get("/stats")
async def get_cache_stats():
    """Get model cache statistics"""
    return model_cache.get_stats()


@router.post("/clear")
async def clear_cache():
    """Clear all cached models"""
    await model_cache.clear()
    return {"message": "Cache cleared successfully"}


@router.post("/preload/{model_name}")
async def preload_model(model_name: str):
    """Preload a model into cache"""
    from app.core.slm_registry import slm_registry
    from app.engines.micro_slm import MicroSLMEngine
    
    micro_slm_info = slm_registry.get_model(model_name)
    if not micro_slm_info:
        return {"error": f"Model {model_name} not found in registry"}
    
    async def load_micro_slm():
        engine = MicroSLMEngine(
            name=model_name,
            model_path=micro_slm_info.model_path
        )
        await engine.initialize()
        return engine
    
    await model_cache.preload(model_name, load_micro_slm)
    
    return {"message": f"Model {model_name} preloaded successfully"}