Optimizing Small Language Models (SLMs) for Real-Time Inference on Edge Devices: Challenges and Best Practices
Practical guide to deploy and optimize small language models for low-latency, power-constrained edge inference with quantization, pruning, and runtime tips.