From 1a9543fadd80e5df817690360ac80da000083064 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Mon, 5 Jan 2026 07:03:34 +0000
Subject: [PATCH] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 unsloth/models/rl.py | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/unsloth/models/rl.py b/unsloth/models/rl.py
index 35a15d03a6..23cbbf0256 100644
--- a/unsloth/models/rl.py
+++ b/unsloth/models/rl.py
@@ -46,10 +46,13 @@ torch_compile_options = {
 # vLLM compatibility shim (TRL expects GuidedDecodingParams even if vLLM doesn't provide it)
 try:
     import vllm.sampling_params as _unsloth_vllm_sp
+
     if not hasattr(_unsloth_vllm_sp, "GuidedDecodingParams"):
+
         class GuidedDecodingParams:
             def __init__(self, **kwargs):
                 self.kwargs = kwargs
+
         _unsloth_vllm_sp.GuidedDecodingParams = GuidedDecodingParams
 except Exception:
     pass
@@ -355,6 +358,7 @@ class Unsloth{RLTrainer_name}(_Unsloth{RLTrainer_name}):
 pass
 '''
 
+
 def _wrap_grpo_generate_and_score(trainer_cls):
     if not hasattr(trainer_cls, "_generate_and_score_completions"):
         return
@@ -367,7 +371,11 @@ def _wrap_grpo_generate_and_score(trainer_cls):
         try:
             return original(self, *args, **kwargs)
         finally:
-            if was_training is False and hasattr(self, "model") and hasattr(self.model, "for_inference"):
+            if (
+                was_training is False
+                and hasattr(self, "model")
+                and hasattr(self.model, "for_inference")
+            ):
                 try:
                     self.model.for_inference()
                 except Exception: