Tagged: model-serving

7 articles

Latest

BERT Variants Get 2x Faster in Hugging Face Transformers, Even on TPUs

DeepFloyd IF Needs 2 GPUs—BentoML Shows How to Actually Serve It

Rocket Money ditched its regex system for Hugging Face to handle 100M monthly transactions

Stop benchmarking LLMs wrong: why time.perf_counter() alone is a trap

Why your LLM's GPU is mostly idle—and how continuous batching fixes it

Hugging Face transformers now runs at native vLLM speed with zero porting

Hugging Face's TGI now serves 30 LoRA models from a single GPU deployment