Add uv scripts headers

examples/scripts/alignprop.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "Pillow>=9.4.0",
+# ]
+# ///
+
 """
 Total Batch size = 128 = 4 (num_gpus) * 8 (per_device_batch) * 4 (accumulation steps)
 Feel free to reduce batch size or increasing truncated_rand_backprop_min to a higher value to reduce memory usage.

examples/scripts/bco.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "peft",
+# ]
+# ///
+
 """
 Run the BCO training script with the commands below. In general, the optimal configuration for BCO will be similar to that of KTO.
 

examples/scripts/cpo.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "peft",
+# ]
+# ///
+
 """
 Run the CPO training script with the following command with some example arguments.
 In general, the optimal configuration for CPO will be similar to that of DPO:

examples/scripts/ddpo.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 """
 python examples/scripts/ddpo.py \
     --num_epochs=200 \

examples/scripts/dpo_online.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "peft",
+# ]
+# ///
+
 """
 Usage:
 

examples/scripts/dpo_vlm.py

@@ -12,6 +12,14 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "peft",
+#     "Pillow>=9.4.0",
+# ]
+# ///
+
 """
 Without dataset streaming:
 

examples/scripts/evals/judge_tldr.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "vllm",
+# ]
+# ///
+
 from dataclasses import dataclass, field
 from typing import Optional
 

examples/scripts/gkd.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "peft",
+# ]
+# ///
+
 """
 # Full training:
 python examples/scripts/gkd.py \

examples/scripts/grpo_vlm.py

@@ -12,6 +12,15 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "peft",
+#     "math-verify",
+#     "latex2sympy2_extended",
+# ]
+# ///
+
 """
 pip install math_verify
 

examples/scripts/kto.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "peft",
+# ]
+# ///
+
 """
 Run the KTO training script with the commands below. In general, the optimal configuration for KTO will be similar to that of DPO.
 

examples/scripts/nash_md.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 """
 Usage:
 

examples/scripts/orpo.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "peft",
+# ]
+# ///
+
 """
 Run the ORPO training script with the following command with some example arguments.
 In general, the optimal configuration for ORPO will be similar to that of DPO without the need for a reference model:

examples/scripts/ppo/ppo.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "peft",
+# ]
+# ///
+
 import shutil
 
 import torch

examples/scripts/ppo/ppo_tldr.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "peft",
+# ]
+# ///
+
 import shutil
 
 import torch

examples/scripts/prm.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 """
 Full training:
 python examples/scripts/prm.py \

examples/scripts/reward_modeling.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 """
 Full training:
 python examples/scripts/reward_modeling.py \

examples/scripts/rloo/rloo.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 import shutil
 
 from accelerate import PartialState

examples/scripts/rloo/rloo_tldr.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 import shutil
 
 from accelerate import PartialState

examples/scripts/sft_gemma3.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 """
 Train Gemma-3 on the Codeforces COTS dataset.
 

examples/scripts/sft_video_llm.py

@@ -12,6 +12,15 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "peft",
+#     "wandb",
+#     "qwen-vl-utils",
+# ]
+# ///
+
 """
 Example usage:
 accelerate launch \

examples/scripts/sft_vlm.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "Pillow>=9.4.0",
+# ]
+# ///
+
 """
 pip install pillow
 

examples/scripts/sft_vlm_gemma3.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "Pillow>=9.4.0",
+# ]
+# ///
+
 """
 Train Gemma-3 on the HuggingFaceH4/llava-instruct-mix-vsft dataset (single-image).
 

examples/scripts/sft_vlm_smol_vlm.py

@@ -12,6 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+#     "Pillow>=9.4.0",
+# ]
+# ///
+
 """
 pip install pillow
 

examples/scripts/xpo.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 """
 Usage:
 

trl/scripts/dpo.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 """
 # Full training
 ```bash

trl/scripts/env.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 import os
 import platform
 from importlib.metadata import version

trl/scripts/grpo.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 import argparse
 import importlib
 import os

trl/scripts/kto.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 """
 Run the KTO training script with the commands below. In general, the optimal configuration for KTO will be similar to
 that of DPO.

trl/scripts/sft.py

@@ -12,6 +12,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# /// script
+# dependencies = [
+#     "trl @ git+https://github.com/huggingface/trl.git",
+# ]
+# ///
+
 """
 # Full training
 ```