9B: 41/8.4 5GB tient; 30B/35B OOM. Plafond=9B+4B parallele, MoE hors-jeu
This commit is contained in:
parent
47e75d74a4
commit
f82fa61281
|
|
@ -78,3 +78,6 @@ HMX on/off=54.7 identique, b1024=54.7 -> GDN scalaire = ancre (pas matmul). Plaf
|
|||
|
||||
## 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.
|
||||
|
||||
## 9B + verdict gros modeles (24/05)
|
||||
Qwen3.5-9B-Q4_0: prefill HTP 41 / decode CPU 8.4 / 5GB. 30B-A3B Q3 13GB=OOM (11.7 dispo). 35B FP8/Int4 non-GGUF+ne tient pas. Plafond tablette: 9B seul ou 9B+4B parallele. MoE hors-jeu. Decode BW: 4B=15,9B=8.4. Fix GDN s applique au 9B.
|
||||
|
|
|
|||
Loading…
Reference in New Issue