service-level objectives

4 papers

• ElasticMM: Efficient Multimodal LLMs Serving with Elastic Multimodal Parallelism
• ElasticMM: Efficient Multimodal LLMs Serving with Elastic Multimodal Parallelism
• HyGen: Efficient LLM Serving via Elastic Online-Offline Request Co-location
• NestedFP: High-Performance, Memory-Efficient Dual-Precision Floating Point Support for LLMs