diff --git a/tests/integration/defs/accuracy/test_llm_api_pytorch.py b/tests/integration/defs/accuracy/test_llm_api_pytorch.py
index 71c351ecf36b..cefe07578497 100644
--- a/tests/integration/defs/accuracy/test_llm_api_pytorch.py
+++ b/tests/integration/defs/accuracy/test_llm_api_pytorch.py
@@ -1400,9 +1400,12 @@ def test_auto_dtype(self):
 
     def test_fp8_prequantized(self):
         # Disabling kv cache reuse as a WAR to deal with gaps in kernel support for Gemma3's non-inclusive sliding window size.
+        # Cap KV cache at 50% of free memory to leave PyTorch
+        # activation headroom during MMLU (nvbugs/6701493).
         kv_cache_config = KvCacheConfig(enable_block_reuse=False,
                                         enable_partial_reuse=False,
-                                        dtype="fp8")
+                                        dtype="fp8",
+                                        free_gpu_memory_fraction=0.5)
         # Note: This has only the LLM part quantized. Vision part is in bfloat16.
         prequantized_model_path = f"{llm_models_root()}/gemma/gemma-3-27b-it-fp8/"
         with LLM(prequantized_model_path,