Reduce small model context and rebalance GPU layers

2026-08-25 00:59:41 -07:00
parent 7a17984dbc
commit 339325719e
7 changed files with 38 additions and 35 deletions
@@ -20,7 +20,7 @@ graph LR
        B[upstream llama.cpp b10121
 frank-default-small, port 7770]
        C[Ornith-1.5-35B-A3B-CRACK Q6_K
-native MTP, selected experts offloaded via override, 589824 ctx]
+native MTP, selected experts offloaded via override, 491520 ctx]
        A --> B --> C
    end