releasing 1.6.2 [rebase & merge]

.github/workflows/ci-tests-gpu.yml

@@ -26,7 +26,7 @@ jobs:
   run-gpu-tests:
     name: Testing
     if: github.event_name == 'push' || !github.event.pull_request.draft
-    timeout-minutes: 35
+    timeout-minutes: 40
     runs-on: Roboflow-GPU-VM-Runner
     steps:
       - name: 🖥️ Print GPU information

CHANGELOG.md

@@ -9,6 +9,23 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ---
 
+## [1.6.2] — 2026-03-27
+
+### Added
+
+- `RFDETR.predict(shape=...)` — optional `(height, width)` tuple overrides the default square inference resolution; useful when matching a non-square ONNX export. Both dimensions must be positive integers divisible by `patch_size × num_windows` as determined by the model configuration. (#866)
+
+### Changed
+
+- `ModelConfig.device` and `RFDETR.train(device=...)` now accept `torch.device` objects and indexed device strings such as `"cuda:0"`. Values are normalized to canonical torch-style strings. `RFDETR.train()` warns when an unmapped device type is passed to PyTorch Lightning auto-detection. (#872)
+
+### Fixed
+
+- Fixed ONNX export ignoring an explicit `patch_size` argument: `export()` and `predict()` now resolve `patch_size` from `model_config` by default, validate it strictly (positive integer, not bool), and enforce that `(H, W)` dimensions are divisible by `patch_size × num_windows`. (#876)
+- Fixed ONNX export for models with dynamic batch dimensions — replaced `H_.expand(N_)` with `torch.full` for Python-int spatial dims to eliminate tracer failures. (#871)
+
+---
+
 ## [1.6.1] — 2026-03-25
 
 ### Deprecated

docs/tutorials/index.md

@@ -70,6 +70,15 @@ Use the resources below to learn how to train, build with, and deploy RF-DETR mo
 
     [:octicons-arrow-right-24: Read the tutorial](https://blog.roboflow.com/how-to-deploy-rf-detr-to-an-nvidia-jetson/)
 
+- **Deploy RF-DETR with LitServe [Lightning AI Studio]**
+
+    ---
+
+    ![](https://blog.roboflow.com/content/images/size/w1000/format/webp/2025/03/img-blog-nycerebro--2--min.png)
+    Learn how to deploy RF-DETR as a scalable inference server using LitServe, the AI model serving framework from Lightning AI.
+
+    [:octicons-arrow-right-24: Open the Studio](https://lightning.ai/bhimrajyadav/studios/deploy-rf-detr-a-sota-real-time-object-detection-model-using-litserve)
+
 - **Train and Deploy RF-DETR Models with Roboflow**
 
     ---

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "rfdetr"
-version = "1.6.1"
+version = "1.6.2"
 description = "RF-DETR"
 readme = "README.md"
 authors = [

src/rfdetr/config.py

@@ -68,7 +68,8 @@ class ModelConfig(BaseConfig):
     amp: bool = True
     num_classes: int = 90
     pretrain_weights: Optional[str] = None
-    device: Literal["cpu", "cuda", "mps"] = DEVICE
+    # torch.device values are accepted at validation time and normalized to string.
+    device: str = DEVICE
     resolution: int
     group_detr: int = 13
     gradient_checkpointing: bool = False
@@ -94,6 +95,32 @@ def expand_path(cls, v: Optional[str]) -> Optional[str]:
             return v
         return os.path.realpath(os.path.expanduser(v))
 
+    @field_validator("device", mode="before")
+    @classmethod
+    def _normalize_device(cls, v: Any) -> str:
+        """Normalize supported device inputs to a canonical torch-style string.
+
+        Args:
+            v: Device specifier provided by callers. Supported values are
+                ``str`` (for example ``"cpu"``, ``"cuda"``, ``"cuda:1"``)
+                and ``torch.device``.
+
+        Returns:
+            Canonical string form of the parsed device (for example ``"cuda:1"``).
+
+        Raises:
+            ValueError: If a string value cannot be parsed as a valid torch device.
+            ValueError: If ``v`` is not a string or ``torch.device``.
+        """
+        if isinstance(v, torch.device):
+            return str(v)
+        if isinstance(v, str):
+            try:
+                return str(torch.device(v))
+            except (TypeError, ValueError, RuntimeError) as exc:
+                raise ValueError(f"Invalid device specifier: {v!r}.") from exc
+        raise ValueError("device must be a string or torch.device.")
+
 
 class RFDETRBaseConfig(ModelConfig):
     """