diff --git a/evaluation/EDA/scripts/run_infer.sh b/evaluation/EDA/scripts/run_infer.sh
index 06107a7b2c..0e0d58ebdb 100755
--- a/evaluation/EDA/scripts/run_infer.sh
+++ b/evaluation/EDA/scripts/run_infer.sh
@@ -57,5 +57,3 @@ fi
 # Run the command
 echo $COMMAND
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/agent_bench/scripts/run_infer.sh b/evaluation/agent_bench/scripts/run_infer.sh
index 11c0105d26..64023973af 100755
--- a/evaluation/agent_bench/scripts/run_infer.sh
+++ b/evaluation/agent_bench/scripts/run_infer.sh
@@ -36,5 +36,3 @@ fi
 
 # Run the command
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/biocoder/scripts/run_infer.sh b/evaluation/biocoder/scripts/run_infer.sh
index 584529d51e..85c57e88b8 100755
--- a/evaluation/biocoder/scripts/run_infer.sh
+++ b/evaluation/biocoder/scripts/run_infer.sh
@@ -39,5 +39,3 @@ fi
 # Run the command
 echo $COMMAND
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/bird/scripts/run_infer.sh b/evaluation/bird/scripts/run_infer.sh
index b042c715ee..7776d1f782 100755
--- a/evaluation/bird/scripts/run_infer.sh
+++ b/evaluation/bird/scripts/run_infer.sh
@@ -36,5 +36,3 @@ fi
 
 # Run the command
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/gaia/scripts/run_infer.sh b/evaluation/gaia/scripts/run_infer.sh
index 2dcba3694e..9f50f9f444 100755
--- a/evaluation/gaia/scripts/run_infer.sh
+++ b/evaluation/gaia/scripts/run_infer.sh
@@ -47,5 +47,3 @@ fi
 
 # Run the command
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/gorilla/scripts/run_infer.sh b/evaluation/gorilla/scripts/run_infer.sh
index 38b37b0b51..468b89c220 100644
--- a/evaluation/gorilla/scripts/run_infer.sh
+++ b/evaluation/gorilla/scripts/run_infer.sh
@@ -45,5 +45,3 @@ fi
 
 # Run the command
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/gpqa/scripts/run_infer.sh b/evaluation/gpqa/scripts/run_infer.sh
index 2b004ff577..43326a30cd 100755
--- a/evaluation/gpqa/scripts/run_infer.sh
+++ b/evaluation/gpqa/scripts/run_infer.sh
@@ -44,5 +44,3 @@ fi
 
 # Run the command
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/humanevalfix/scripts/run_infer.sh b/evaluation/humanevalfix/scripts/run_infer.sh
index ca5abaeb05..c61b1be309 100755
--- a/evaluation/humanevalfix/scripts/run_infer.sh
+++ b/evaluation/humanevalfix/scripts/run_infer.sh
@@ -74,5 +74,3 @@ fi
 
 # Run the command
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/logic_reasoning/scripts/run_infer.sh b/evaluation/logic_reasoning/scripts/run_infer.sh
index c32e4c42c8..5c7748b3c9 100755
--- a/evaluation/logic_reasoning/scripts/run_infer.sh
+++ b/evaluation/logic_reasoning/scripts/run_infer.sh
@@ -40,5 +40,3 @@ fi
 
 # Run the command
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/miniwob/scripts/run_infer.sh b/evaluation/miniwob/scripts/run_infer.sh
index 36278b8969..b1dc24a147 100755
--- a/evaluation/miniwob/scripts/run_infer.sh
+++ b/evaluation/miniwob/scripts/run_infer.sh
@@ -46,5 +46,3 @@ fi
 
 # Run the command
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/mint/scripts/run_infer.sh b/evaluation/mint/scripts/run_infer.sh
index 5cac4beac8..6d6a6510b0 100644
--- a/evaluation/mint/scripts/run_infer.sh
+++ b/evaluation/mint/scripts/run_infer.sh
@@ -42,5 +42,3 @@ fi
 
 # Run the command
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/ml_bench/scripts/run_infer.sh b/evaluation/ml_bench/scripts/run_infer.sh
index 9909276e7d..98bc528c63 100755
--- a/evaluation/ml_bench/scripts/run_infer.sh
+++ b/evaluation/ml_bench/scripts/run_infer.sh
@@ -46,5 +46,3 @@ fi
 
 # Run the command
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/toolqa/scripts/run_infer.sh b/evaluation/toolqa/scripts/run_infer.sh
index 6035c71bd9..00d015c338 100644
--- a/evaluation/toolqa/scripts/run_infer.sh
+++ b/evaluation/toolqa/scripts/run_infer.sh
@@ -61,5 +61,3 @@ fi
 
 # Run the command
 eval $COMMAND
-
-checkout_original_branch
diff --git a/evaluation/utils/version_control.sh b/evaluation/utils/version_control.sh
index 38c2dd370f..f58f9bdce9 100644
--- a/evaluation/utils/version_control.sh
+++ b/evaluation/utils/version_control.sh
@@ -1,8 +1,16 @@
 checkout_eval_branch() {
     if [ -z "$COMMIT_HASH" ]; then
         echo "Commit hash not specified, use current git commit"
+        build_sandbox
         return 0
     fi
+
+    if git diff --quiet $COMMIT_HASH HEAD; then
+        echo "The given hash is equivalent to the current HEAD"
+        build_sandbox
+        return 0
+    fi
+
     echo "Start to checkout opendevin version to $COMMIT_HASH, but keep current evaluation harness"
     if ! git diff-index --quiet HEAD --; then
         echo "There are uncommitted changes, please stash or commit them first"
@@ -15,8 +23,20 @@ checkout_eval_branch() {
         echo "Failed to check out to $COMMIT_HASH"
         exit 1
     fi
+
     echo "Revert changes in evaluation folder"
     git checkout $current_branch -- evaluation
+
+    # Trap the EXIT signal to checkout original branch
+    trap checkout_original_branch EXIT
+
+    build_sandbox
+}
+
+build_sandbox() {
+    echo "Build sandbox locally"
+    docker build -t eval-sandbox -f containers/sandbox/Dockerfile /tmp
+    export SANDBOX_CONTAINER_IMAGE="eval-sandbox"
 }
 
 checkout_original_branch() {
diff --git a/evaluation/webarena/scripts/run_infer.sh b/evaluation/webarena/scripts/run_infer.sh
index 1fa9ba13b2..ec6981c9e7 100755
--- a/evaluation/webarena/scripts/run_infer.sh
+++ b/evaluation/webarena/scripts/run_infer.sh
@@ -44,5 +44,3 @@ fi
 
 # Run the command
 eval $COMMAND
-
-checkout_original_branch