EdgeTP: Enabling Distributed Full-size Large Language Model Inference on Edge Devices with Tensor Parallelism

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: EdgeTP: Enabling Distributed Full-size Large Language Model Inference on Edge Devices with Tensor Parallelism
Συγγραφείς: Cui, Laizhong, Guo, Yi, Peng, Weixuan, Huang, Yaodong
Πηγή: 2026 IEEE 46th International Conference on Distributed Computing Systems (ICDCS) ICDCS Distributed Computing Systems (ICDCS), 2026 IEEE 46th International Conference on. :1378-1388 Jun, 2026
Relation: 2026 IEEE 46th International Conference on Distributed Computing Systems (ICDCS)
Βάση Δεδομένων: IEEE Xplore Digital Library
Περιγραφή
ISBN:9798319529794
9798319529800
ISSN:25758411
10636927
DOI:10.1109/2575-8411.2026.00134