diff --git a/.env.example b/.env.example
index 13ae041..8ade668 100644
--- a/.env.example
+++ b/.env.example
@@ -7,6 +7,8 @@ VLLM_API_KEY=
 MODEL_DIR=/data/models/halt95/Swift1.5-Qwen3.8-Flash-Next-W4A16-Merlin
 MODEL_MOUNT_PATH=/models/Qwen3.8-Flash-Next
 MODEL_NAME=Hinata
+# The FP8 KV scale sidecar, a host path. Default: $MODEL_DIR/qsa_kv_scales.json.
+# SCALES=
 
 # ── image and container ──────────────────────────────────────────────────────
 IMAGE=qwen38-flash-next
diff --git a/docker-compose.yml b/docker-compose.yml
index d2433a6..2a18dd2 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -28,7 +28,7 @@ services:
     environment:
       MODEL_NAME: ${MODEL_NAME:-Hinata}
       PORT: 8000
-      SCALES: ${SCALES:-}
+      SCALES: /opt/vllm/qsa_kv_scales.json
       VLLM_API_KEY: ${VLLM_API_KEY:-}
       PLE_HOME: ${PLE_HOME:-0}
       CUDA_VISIBLE_DEVICES: ${CUDA_VISIBLE_DEVICES:-0,1,2,3}
@@ -45,6 +45,7 @@ services:
       - "${MODEL_DIR:?Set MODEL_DIR in .env}:${MODEL_MOUNT_PATH:-/models/Qwen3.8-Flash-Next}:ro"
       - ${ABLITERATION_DIR:-./abliteration}:/opt/qwen38/abliteration:ro
       - ${CACHE_HOST_DIR:-./.vllm-cache}:/cache
+      - ${SCALES:-${MODEL_DIR}/qsa_kv_scales.json}:/opt/vllm/qsa_kv_scales.json:ro
       - ./start.sh:/opt/vllm/start.sh:ro
     restart: unless-stopped
     stop_grace_period: 20s