SmallThinker: A Family of Efficient Large Language Models Natively Trained for Local Deployment Paper • 2507.20984 • Published Jul 28, 2025 • 58
Turbo Sparse: Achieving LLM SOTA Performance with Minimal Activated Parameters Paper • 2406.05955 • Published Jun 10, 2024 • 27
PowerInfer-2: Fast Large Language Model Inference on a Smartphone Paper • 2406.06282 • Published Jun 10, 2024 • 39
PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU Paper • 2312.12456 • Published Dec 16, 2023 • 45
