diff --git a/CLAUDE.md b/CLAUDE.md index 8d6a160..a0e26bd 100644 --- a/CLAUDE.md +++ b/CLAUDE.md @@ -75,3 +75,6 @@ Qwen3-4B Engine: prefill HTP 103 / decode CPU 17. Qwen3.5-4B: 55/15. ExecuTorch ## Prefill 3.5: leviers gratuits satures HMX on/off=54.7 identique, b1024=54.7 -> GDN scalaire = ancre (pas matmul). Plafond=~103 (dense). Levier unique = GDN HVX chunkwise (semaines, bit-exact frontiere). state f16 modere. decode 15 fige (BW). B-2 phase2 = HVX. + +## Plafond prefill: HVX deja actif, inutile (55.5=55.9 scalar=hvx) +Mur = dependance serielle 512tok, pas calcul. HVX inner-dot ~0 gain. Cap 55 = recurrence. Seul levier = chunkwise GDN (delta rule matricielle->HMX->~100), changement algo, semaines, bit-exact frontiere. Aucun gain rapide. decode 15 fige.