diff --git a/crates/ml-ppo/src/lib.rs b/crates/ml-ppo/src/lib.rs
index a77a624ec..aea5df64b 100644
--- a/crates/ml-ppo/src/lib.rs
+++ b/crates/ml-ppo/src/lib.rs
@@ -20,7 +20,7 @@
 #![allow(unsafe_code)] // Required for CUDA kernel launches and cuBLAS FFI
 
 // Re-export shared modules from ml-core for convenience
-pub use ml_core::cuda_compat;
+pub use ml_core::cuda_compile;
 
 pub mod adaptive_entropy;
 pub mod continuous_policy;
diff --git a/crates/ml/src/trainers/tft/model.rs b/crates/ml/src/trainers/tft/model.rs
index 27df9aaa9..5afc509ee 100644
--- a/crates/ml/src/trainers/tft/model.rs
+++ b/crates/ml/src/trainers/tft/model.rs
@@ -2,7 +2,7 @@
 //!
 //! This module provides a trait-based abstraction for the TFT model used by TFTTrainer.
 
-use ml_core::cuda_autograd::GpuTensor;
+use ml_core::cuda_autograd::StreamTensor;
 
 use crate::tft::{TFTConfig, TemporalFusionTransformer};
 use crate::MLError;
@@ -21,11 +21,11 @@ pub trait TFTModel: Send + Sync {
     /// * Quantile predictions [batch, horizon, num_quantiles]
     fn forward(
         &mut self,
-        static_features: &GpuTensor,
-        historical_ts: &GpuTensor,
-        future_ts: &GpuTensor,
+        static_features: &StreamTensor,
+        historical_ts: &StreamTensor,
+        future_ts: &StreamTensor,
         use_checkpointing: bool,
-    ) -> Result<GpuTensor, MLError>;
+    ) -> Result<StreamTensor, MLError>;
 
     /// Get configuration
     fn get_config(&self) -> &TFTConfig;
@@ -39,11 +39,11 @@ pub trait TFTModel: Send + Sync {
 impl TFTModel for TemporalFusionTransformer {
     fn forward(
         &mut self,
-        static_features: &GpuTensor,
-        historical_ts: &GpuTensor,
-        future_ts: &GpuTensor,
+        static_features: &StreamTensor,
+        historical_ts: &StreamTensor,
+        future_ts: &StreamTensor,
         use_checkpointing: bool,
-    ) -> Result<GpuTensor, MLError> {
+    ) -> Result<StreamTensor, MLError> {
         self.forward_with_checkpointing(
             static_features,
             historical_ts,