diff --git a/Dockerfile.cuda b/Dockerfile.cuda
index ac72a88..89ef32d 100644
--- a/Dockerfile.cuda
+++ b/Dockerfile.cuda
@@ -22,6 +22,8 @@ ENV THREADS=4
 
 ENV MODEL="/models/codegen-2B-multi-ggml-4bit-quant.bin"
 
+ENV BATCHSIZE=64
+
 COPY ./run.sh /app/
 
 EXPOSE 18080
diff --git a/Dockerfile.default b/Dockerfile.default
index bad770b..d6db39c 100644
--- a/Dockerfile.default
+++ b/Dockerfile.default
@@ -22,6 +22,8 @@ ENV THREADS=4
 
 ENV MODEL="/models/codegen-2B-multi-ggml-4bit-quant.bin"
 
+ENV BATCHSIZE=8
+
 COPY ./run.sh /app/
 
 EXPOSE 18080
diff --git a/run.sh b/run.sh
index 8db96c8..ef1339f 100755
--- a/run.sh
+++ b/run.sh
@@ -1,3 +1,3 @@
 #!/bin/sh
 
-/app/codegen-serve -t $THREADS -m $MODEL
\ No newline at end of file
+/app/codegen-serve -t $THREADS -m $MODEL -b $BATCHSIZE
\ No newline at end of file