Tyler Griggs

MoE-L <scp>ightning</scp> : High-Throughput MoE Inference on Memory-constrained GPUs

Shiyi Cao, Shu Liu, Tyler Griggs, Peter Schafhalter, Xiaoxuan Liu, Ying Sheng, Joseph E. Gonzalez, Matei Zaharia, Ion Stoica (2025). MoE-L <scp>ightning</scp> : High-Throughput MoE Inference on Memory-constrained GPUs. , DOI: https://doi.org/10.1145/3669940.3707267.