Kazeia-engine/docs
Richard Loyer dd29169672 GDN chunkwise P1.2: VTCM state prefill=neutre, prouve compute-bound 3S2/tok -> gain=WY matmul (palier dur). oracle 28/28, pp98 2026-05-26 16:30:08 +02:00
..
01_fix_prefill_overflow.md Chantier 1 clos: prefill pp2048 sain (46t/s), hang=cli only hors JNI 2026-05-24 17:16:44 +02:00
02_cont_overhead.md Chantier 2 clos: CONT low-ROI (opsync gonfle, reshape structurel) -> 3 prioritaire 2026-05-24 17:17:24 +02:00
03_gdn_prefill_htp.md FIX GDN prefill HTP: GQA broadcast h%heads, charabia resolu, pp512 58 t/s (+16%) 2026-05-24 17:57:20 +02:00
04_gdn_chunkwise.md GDN chunkwise P1.2: VTCM state prefill=neutre, prouve compute-bound 3S2/tok -> gain=WY matmul (palier dur). oracle 28/28, pp98 2026-05-26 16:30:08 +02:00