From ed7105dba74e6035cdf53a9cd8a499247f314a91 Mon Sep 17 00:00:00 2001
From: NanoCode012 <nano@axolotl.ai>
Date: Tue, 10 Feb 2026 17:52:09 +0700
Subject: [PATCH] fix: GRPO config not accept max_prompt_length (#3390) [skip
 ci]

---
 src/axolotl/core/trainers/grpo/__init__.py | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/axolotl/core/trainers/grpo/__init__.py b/src/axolotl/core/trainers/grpo/__init__.py
index e611b96ea..8b295e537 100644
--- a/src/axolotl/core/trainers/grpo/__init__.py
+++ b/src/axolotl/core/trainers/grpo/__init__.py
@@ -164,7 +164,12 @@ class GRPOStrategy:
 
     @classmethod
     def get_blocklist_args_kwargs(cls) -> list[str]:
-        return ["dataset_num_proc", "max_length", "include_tokens_per_second"]
+        return [
+            "dataset_num_proc",
+            "max_length",
+            "include_tokens_per_second",
+            "max_prompt_length",
+        ]
 
     @classmethod
     def get_reward_func(cls, reward_func_fqn: str) -> RewardFunc: