diff --git a/CLAUDE.md b/CLAUDE.md index 66c38d4..5422b37 100644 --- a/CLAUDE.md +++ b/CLAUDE.md @@ -90,3 +90,6 @@ Speaker 9B 5GB + Thinker Guard-4B 2.5GB = 7.5GB co-resident, 8GB libre. Guard-8B ## Cmp 4B vs 9B: signal partiel 9B analyse + riche, 4B quasi-vide. Les 2 comprennent FR. Mais cli=decode HTP 0.6t/s + reasoning anglais, /no_think ignore -> eval FR propre IMPOSSIBLE en shell. Reco 9B qualite, 4B RAM. Vrai cmp = JNI (decode CPU+template). Prouve sur 4 runs. + +## Fix moteur cascade: --reasoning-budget 0 (24/05) +BUG=enable_thinking default true -> /no_think ignore -> ramble anglais. FIX: --reasoning-budget 0 (arg.cpp:3061). Prouve: 4B->FR bref coherent "Je suis desole que tu te sens ainsi". Batch shell A/B reste impossible (HTP decode 0.6, EOS variable). Vrai cmp 4B vs 9B = JNI: decode CPU+template+budget0. Engine OK, cascade tient RAM.