Pipeline-parallel LLM inference across GPUs on separate machines

(github.com) by ngaut | Jun 19, 2026 | 0 comments on HN