▲ 0 ▼ c/hackernews@lemmy.bestiver.se · rssbot@lemmy.bestiver.se · 18 Nov 2026 02:30 Cerebras Disaggregated Inference #hackernews Cerebras Launches Disaggregated Inference Cerebras pairs AMD GPUs for prefill with wafer-scale decode to deliver 5× more tokens per megawatt without reducing token generation speed. www.cerebras.ai check post at origin Comments
0 comments
log in to comment.