diff --git a/.env.example b/.env.example
index 4c485b8..1b5dba9 100644
--- a/.env.example
+++ b/.env.example
@@ -5,7 +5,7 @@ LLAMA_MODEL_FILE=Qwen3.8-27B-UD-Q4_K_XL.gguf
 LLAMA_GPU_LAYERS=999
 LLAMA_CTX_SIZE=262144
 LLAMA_PARALLEL=2
-LLAMA_CACHE_RAM=24576
+LLAMA_CACHE_RAM=8192
 LLAMA_KV_CACHE_TYPE=q8_0
 
 # Blank = speculative decoding off; see docs/architecture.md#llama-server