Work

GPU Efficiency Without Compromise: LLM Inference Optimization Through Dynamic Batching, KV-Cache Management, and Quantization

Year: 2026

Type: preprint

Source: OSF Preprints (OSF Preprints)

Author: Bhavin Kotak

Language: English

Topic: Parallel Computing and Optimization Techniques

Subfield: Hardware and Architecture

Field: Computer Science

Domain: Physical Sciences

Open Access status: closed

OpenAlex ID: W7127225950

Cites:

Cited by:

Related to: 10