Attention-FFN Disaggregationが真剣に考えられているということも知らなかったので、本当に勉強になりますね
reactive:inference-cost-optimization · Kazuki Fujii (@kazukifujii) · 2026-06-27
(No summary yet for this item — extraction summaries are still backfilling.)
reactive:inference-cost-optimization · Kazuki Fujii (@kazukifujii) · 2026-06-27
(No summary yet for this item — extraction summaries are still backfilling.)