fixup local-llama service

local-llama.sh

@@ -1,5 +1,8 @@
 #!/bin/bash
 
+export PATH=/usr/local/cuda-13.3/bin${PATH:+:${PATH}}
+export LD_LIBRARY_PATH=/usr/local/cuda-13.3/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
+
 set -euo pipefail
 
 /home/bazzite/Документы/llama-cpp-turboquant/build/bin/llama-server \
@@ -8,7 +11,7 @@ set -euo pipefail
   --port 8080 \
   -ngl "${N_GPU_LAYERS}" \
   --n-cpu-moe "${N_CPU_MOE}" \
-  --spec-type draft-mtp --spec-draft-n-max 3 \
+  --spec-type draft-mtp --spec-draft-n-max 2 \
   -c "${CTX_SIZE}" \
   -np 1 \
   -fa on \