From 9675280760f981fa6a1b2c7301becacae0f14e73 Mon Sep 17 00:00:00 2001
From: Richard Loyer <richard.loyer@c42.lu>
Date: Sun, 24 May 2026 22:12:47 +0200
Subject: [PATCH] GDN chunkwise: HMX fp16 bloque bit-exact, HVX qf32 3-5sem;
 reste 55/15 CPU exact

---
 CLAUDE.md | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/CLAUDE.md b/CLAUDE.md
index e2263c2..06b3974 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -102,3 +102,6 @@ BUG=enable_thinking default true -> /no_think ignore -> ramble anglais. FIX: --r
 
 ## TTS statut 24/05
 Talker=qwen3 GGUF chargeable engine (poids OK). Decoder RTF0.96 chraac valide histo. MAIS audio bout-en-bout sur engine NON valide: bench standalone segfault/libomp perm; vrai TTS=JNI Talker->codes->CP->decoder (integration app). GDN chunkwise: agent en cours. Optim sure: release Thinker -2.5GB, KV f16.
+
+## GDN chunkwise verdict
+Chunkwise batchable, 55->100 credible MAIS HMX accumulateur FP16 -> bit-exact impossible. Voie=HVX qf32 3-5sem frontiere. Prefill GDN-HTP encore not-bit-correct (2287 default OFF), decode CPU bit-exact. Reco: rester 55/15, gain pas worth (4s masques). Optim sure=release Thinker+KV f16.