diff --git a/config/gpu/rtx3050.toml b/config/gpu/rtx3050.toml
index 5c2fa43f3..1c810abef 100644
--- a/config/gpu/rtx3050.toml
+++ b/config/gpu/rtx3050.toml
@@ -1,6 +1,6 @@
 # RTX 3050 Ti (4GB VRAM) -- minimal footprint
 [training]
-batch_size = 128
+batch_size = 64
 num_atoms = 11
 buffer_size = 0  # 0 = auto from VRAM
 hidden_dim_base = 256
diff --git a/crates/ml/src/training_profile.rs b/crates/ml/src/training_profile.rs
index 4cfdc9174..82d455986 100644
--- a/crates/ml/src/training_profile.rs
+++ b/crates/ml/src/training_profile.rs
@@ -670,9 +670,8 @@ impl DqnTrainingProfile {
             if let Some(v) = t.epochs {
                 hp.epochs = v;
             }
-            if let Some(v) = t.batch_size {
-                hp.batch_size = v;
-            }
+            // batch_size: NOT applied from training profile.
+            // GPU profile (config/gpu/*.toml) is the sole authority for batch_size.
             if let Some(v) = t.learning_rate {
                 hp.learning_rate = v;
             }