You Kaichao: Open-Source Infra, vLLM & Model Co-design

Why it matters

A 3-hour interview with vLLM core developer You Kaichao: why 'if vLLM fails, we'll regret it for life', how open-source infra and model co-design reshape inference performance, and where Chinese open inference engines sit in global AI infrastructure.

Key takeaways

• vLLM's mission: making the inference engine an amplifier of model capability.
• The co-design paradigm: models and inference systems designed together, not separately.
• Open-source infra's logic: community-driven vs commercial.
• 'If vLLM fails, we'll regret it for life': the open-source believer's conviction.
• Where Chinese open inference engines sit globally.