diff --git a/giant/config.py b/giant/config.py
index cbcf0e9..7001151 100644
--- a/giant/config.py
+++ b/giant/config.py
@@ -53,14 +53,15 @@ def auto_device() -> torch.device:
     return torch.device("cpu")
 
 
-# Calibration point for estimate_batch_size(training=True): hidden_dim=512,
-# n_blocks=6, batch_size=131072 measured at ~8 GiB VRAM. Activation memory is
-# assumed to scale linearly with batch_size * hidden_dim * n_blocks (the
-# ResBlock stack dominates), so this is a rough estimate rather than a
-# guaranteed bound.
-_REF_BYTES = 8 * 1024**3
-_REF_BATCH_SIZE = 131072
-_REF_HIDDEN_DIM = 512
+# Calibration point for estimate_batch_size(training=True): hidden_dim=1024,
+# n_blocks=6, batch_size=29696 measured at ~7683 MiB VRAM (post-Phase-2
+# architecture, including the Stage-2 secondary decoder and n_sec head).
+# Activation memory is assumed to scale linearly with
+# batch_size * hidden_dim * n_blocks (the ResBlock stack dominates), so this
+# is a rough estimate rather than a guaranteed bound.
+_REF_BYTES = 7683 * 1024**2
+_REF_BATCH_SIZE = 29696
+_REF_HIDDEN_DIM = 1024
 _REF_N_BLOCKS = 6
 
 # Calibration point for estimate_batch_size(training=False): inference has no