SamplingParams

unsloth/models/__init__.py

@@ -20,4 +20,4 @@ from .mistral import FastMistralModel
 from .qwen2   import FastQwen2Model
 from .dpo     import PatchDPOTrainer, PatchKTOTrainer
 from ._utils  import is_bfloat16_supported
-from .rl      import PatchFastRL
+from .rl      import PatchFastRL, vLLMSamplingParams

unsloth/models/rl.py

@@ -44,6 +44,7 @@ torch_compile_options = {
 
 
 def vLLMSamplingParams(**kwargs):
+    from vllm import SamplingParams
     sampling_params = SamplingParams(**kwargs)
     sampling_params._set_kwargs = kwargs
     return sampling_params