[https://nvbugs/6143787][fix] Add kv_cache_config = KvCacheConfig(free_gpu_memory_fraction=0.6) to TestQwen3 (#13852)

Signed-off-by: tensorrt-cicd <90828364+tensorrt-cicd@users.noreply.github.com>

Author: tensorrt-cicd

tests/integration/defs/accuracy/test_llm_api_pytorch_multimodal.py

@@ -480,6 +480,8 @@ class TestQwen3VL(LlmapiAccuracyTestHarness):
         max_tokens=MAX_NUM_TOKENS, truncate_prompt_tokens=MMMU.MAX_INPUT_LEN, stop="<|endoftext|>"
     )
 
+    kv_cache_config = KvCacheConfig(free_gpu_memory_fraction=0.6)
+
     @pytest.mark.parametrize(
         "enable_chunked_prefill,max_num_tokens",
         [
@@ -493,6 +495,7 @@ def test_auto_dtype(self, enable_chunked_prefill, max_num_tokens):
             self.MODEL_PATH,
             enable_chunked_prefill=enable_chunked_prefill,
             max_num_tokens=max_num_tokens,
+            kv_cache_config=self.kv_cache_config,
         ) as llm:
             task = MMMU(self.MODEL_NAME)
             task.evaluate(llm, sampling_params=self.sampling_params)

tests/integration/test_lists/waives.txt

@@ -147,7 +147,6 @@ accuracy/test_llm_api_pytorch_multimodal.py::TestExaone4_5_33B::test_auto_dtype[
 accuracy/test_llm_api_pytorch_multimodal.py::TestGemma3_27BInstruct::test_fp8_prequantized SKIP (https://nvbugs/6215689)
 accuracy/test_llm_api_pytorch_multimodal.py::TestMistralLarge3_675B::test_nvfp4_4gpus[latency_moe_trtllm] SKIP (https://nvbugs/6181383)
 accuracy/test_llm_api_pytorch_multimodal.py::TestNemotron_Nano_12B_V2_VL::test_auto_dtype[forced_chunked_prefill] SKIP (https://nvbugs/6248744)
-accuracy/test_llm_api_pytorch_multimodal.py::TestQwen3VL::test_auto_dtype[forced_chunked_prefill] SKIP (https://nvbugs/6143787)
 accuracy/test_llm_api_pytorch_ray.py::TestLlama3_1_8BInstruct::test_pp2_ray SKIP (https://nvbugs/6094070)
 cpp/test_e2e.py::test_benchmarks[bart-90] SKIP (https://nvbugs/5550689)
 cpp/test_e2e.py::test_benchmarks[gpt-80] SKIP (https://nvbugs/5550689)