1. 2026
  2. Multithreaded Pipeline Parallelism for Distributed LLM Inference Without Source Code Modification

    Koloskov, V. & Pershin, A., 30 Jun 2026, Computational Science and Its Applications – ICCSA 2026 Workshops. Springer Nature, p. 464-480 17 p. Chapter 29. (Computational Science and Its Applications – ICCSA 2026 Workshops; vol. 16763).

    Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

ID: 97883755