OpenAI and Broadcom unveil LLM-optimized inference chip

Re:OpenAI and Broadcom unveil LLM-optimized inference chip by kccqzy on Wednesday June 24, 2026 @04:45PM (#48665393)

Well Google has reduced reliance on Broadcom already. They found a new hardware partner, MediaTek, that’s probably much, much cheaper than Broadcom.

[ Parent | on HN ]

Re:OpenAI and Broadcom unveil LLM-optimized inference chip by mschuster91 on Wednesday June 24, 2026 @06:06PM (#48666176)

> Well Google has reduced reliance on Broadcom already. They found a new hardware partner, MediaTek

Oh dear god. I'm actually feeling sorry for Google at that point. Good luck, you'll need it...

[ Parent | on HN ]

Re:OpenAI and Broadcom unveil LLM-optimized inference chip by kccqzy on Wednesday June 24, 2026 @07:22PM (#48666761)

My hunch is that this change is driven by bean counters.

[ Parent | on HN ]

Re:OpenAI and Broadcom unveil LLM-optimized inference chip by amelius on Thursday June 25, 2026 @07:44AM (#48672005)

Who says Google isn't doing its own designs mostly?

[ Parent | on HN ]

Re:OpenAI and Broadcom unveil LLM-optimized inference chip by kccqzy on Thursday June 25, 2026 @08:40AM (#48672482)

Oh they definitely are. But as a transitional step, replacing Broadcom with MediaTek is probably mostly about cost.

[ Parent | on HN ]

Re:OpenAI and Broadcom unveil LLM-optimized inference chip by rasz on Wednesday June 24, 2026 @09:05PM (#48667503)

MediaTek was spun out of UMC. UMC was a powerhouse of ASIC design.

[ Parent | on HN ]