Synthesis history

The AMD-Red Hat strategic collaboration announcement [^20479] from May 2025 is the primary new development this pass: it reframes AMD's absence from llm-d as a transitional gap within an active partnership rather than a…

Two new dimensions enter the story this pass. First, Red Hat has moved from governance co-donor to enterprise product owner: Red Hat AI 3 ships llm-d integrated into OpenShift AI [^19908][^19906][^19905], making AMD's a…

MLPerf Inference v6.0 results are now officially published by MLCommons [^18663], moving AMD's competitiveness claim from self-described 'breakthrough' to external documentation — but generating split interpretations, w…

IBM Research's donation blog reveals IBM as a co-originator of the llm-d CNCF donation alongside Google, correcting the prior synthesis which attributed it to Google alone and broadening the coalition behind the neutral…

No new fault lines this pass, but several confirmatory details sharpen the existing narrative. AMD's pricing strategy became explicit: the MI355X costs more to build than NVIDIA's competing chip yet sells for significan…

Over five days in mid-May 2026, three separate competitive signals converged: AMD's MI355 GPU achieved 40% lower inference cost than NVIDIA's B200 on the GLM5 architecture just 14 weeks after launch [^7778]; Google adde…