diff --git a/tests/integration/defs/conftest.py b/tests/integration/defs/conftest.py
index 333941b57cf..0962c24b636 100644
--- a/tests/integration/defs/conftest.py
+++ b/tests/integration/defs/conftest.py
@@ -15,6 +15,7 @@
 # -*- coding: utf-8 -*-
 
 import datetime
+import gc
 import os
 import platform
 import re
@@ -2562,4 +2563,5 @@ def torch_empty_cache() -> None:
     Manually empty the torch CUDA cache before each test, to reduce risk of OOM errors.
     """
     if torch.cuda.is_available():
+        gc.collect()
         torch.cuda.empty_cache()
diff --git a/tests/integration/defs/test_e2e.py b/tests/integration/defs/test_e2e.py
index 3defa73aac2..ad4f3e6a621 100644
--- a/tests/integration/defs/test_e2e.py
+++ b/tests/integration/defs/test_e2e.py
@@ -2262,7 +2262,7 @@ def test_ptp_quickstart_advanced_deepseek_r1_w4afp8_8gpus(
 
 @pytest.mark.skip_less_device_memory(80000)
 @pytest.mark.parametrize("model_name,model_path,gpu_count", [
-    ("Llama3.1-70B-BF16", "llama-3.1-model/Meta-Llama-3.1-70B", 2),
+    ("Llama3.1-70B-BF16", "llama-3.1-model/Meta-Llama-3.1-70B", 8),
     ("Mixtral-8x7B-BF16", "Mixtral-8x7B-v0.1", 8),
     pytest.param('Llama3.1-70B-FP8',
                  'llama-3.1-model/Llama-3.1-70B-Instruct-FP8',
@@ -2293,7 +2293,7 @@ def test_ptp_quickstart_advanced_multi_gpus(llm_root, llm_venv, model_name,
         pytest.skip(f"Not enough GPUs for {model_name}")
     example_root = Path(os.path.join(llm_root, "examples", "llm-api"))
     mapping = {
-        "Llama3.1-70B-BF16": 91.0,
+        "Llama3.1-70B-BF16": 24.6,
         "Mixtral-8x7B-BF16": 16.5,
         "Llama3.1-70B-FP8": 58.5,
         "Llama3.1-405B-FP8": 63.2,
diff --git a/tests/integration/test_lists/qa/llm_function_core.txt b/tests/integration/test_lists/qa/llm_function_core.txt
index 0d3985d4fde..d851cae5432 100644
--- a/tests/integration/test_lists/qa/llm_function_core.txt
+++ b/tests/integration/test_lists/qa/llm_function_core.txt
@@ -634,7 +634,7 @@ test_e2e.py::test_ptp_quickstart_advanced[Llama3.1-8B-NVFP4-nvfp4-quantized/Meta
 test_e2e.py::test_ptp_quickstart_advanced[Llama3.2-11B-BF16-llama-3.2-models/Llama-3.2-11B-Vision]
 test_e2e.py::test_ptp_quickstart_advanced[Qwen3-30B-A3B-Qwen3/Qwen3-30B-A3B]
 test_e2e.py::test_ptp_quickstart_advanced_ngram[Llama-3.1-8B-Instruct-llama-3.1-model/Llama-3.1-8B-Instruct]
-test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-BF16-llama-3.1-model/Meta-Llama-3.1-70B-2]
+test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-BF16-llama-3.1-model/Meta-Llama-3.1-70B-8]
 test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-FP8-llama-3.1-model/Llama-3.1-70B-Instruct-FP8-2]
 test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-405B-FP8-llama-3.1-model/Llama-3.1-405B-Instruct-FP8-8]
 test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Mixtral-8x7B-BF16-Mixtral-8x7B-v0.1-8]
diff --git a/tests/integration/test_lists/qa/llm_function_core_sanity.txt b/tests/integration/test_lists/qa/llm_function_core_sanity.txt
index 4ad8bea44c1..fd8b9cc979f 100644
--- a/tests/integration/test_lists/qa/llm_function_core_sanity.txt
+++ b/tests/integration/test_lists/qa/llm_function_core_sanity.txt
@@ -210,7 +210,7 @@ test_e2e.py::test_openai_consistent_chat
 test_e2e.py::test_openai_multi_chat_example
 test_e2e.py::test_ptp_quickstart
 test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-405B-FP8-llama-3.1-model/Llama-3.1-405B-Instruct-FP8-8]
-test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-BF16-llama-3.1-model/Meta-Llama-3.1-70B-2]
+test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-BF16-llama-3.1-model/Meta-Llama-3.1-70B-8]
 test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-FP8-llama-3.1-model/Llama-3.1-70B-Instruct-FP8-2]
 test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Mixtral-8x7B-BF16-Mixtral-8x7B-v0.1-8]
 test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Mixtral-8x7B-NVFP4-nvfp4-quantized/Mixtral-8x7B-Instruct-v0.1-8]
diff --git a/tests/integration/test_lists/qa/llm_function_nim.txt b/tests/integration/test_lists/qa/llm_function_nim.txt
index 1e538f25a59..4e3812ddd06 100644
--- a/tests/integration/test_lists/qa/llm_function_nim.txt
+++ b/tests/integration/test_lists/qa/llm_function_nim.txt
@@ -408,7 +408,7 @@ test_e2e.py::test_ptp_quickstart_advanced[Llama3.1-8B-BF16-llama-3.1-model/Meta-
 test_e2e.py::test_ptp_quickstart_advanced[Llama3.1-8B-FP8-llama-3.1-model/Llama-3.1-8B-Instruct-FP8]
 test_e2e.py::test_ptp_quickstart_advanced[Llama3.1-8B-NVFP4-nvfp4-quantized/Meta-Llama-3.1-8B]
 test_e2e.py::test_ptp_quickstart_advanced[Qwen3-30B-A3B-Qwen3/Qwen3-30B-A3B]
-test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-BF16-llama-3.1-model/Meta-Llama-3.1-70B-2]
+test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-BF16-llama-3.1-model/Meta-Llama-3.1-70B-8]
 test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-FP8-llama-3.1-model/Llama-3.1-70B-Instruct-FP8-2]
 test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-405B-FP8-llama-3.1-model/Llama-3.1-405B-Instruct-FP8-8]
 test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Mixtral-8x7B-BF16-Mixtral-8x7B-v0.1-8]
diff --git a/tests/integration/test_lists/waives.txt b/tests/integration/test_lists/waives.txt
index 4e2bb48655d..6286626159b 100644
--- a/tests/integration/test_lists/waives.txt
+++ b/tests/integration/test_lists/waives.txt
@@ -313,7 +313,6 @@ full:H100/accuracy/test_llm_api_pytorch.py::TestLlama4MaverickInstruct::test_fp8
 full:H100/accuracy/test_llm_api_pytorch.py::TestLlama4MaverickInstruct::test_fp8[tp8-cuda_graph=True] SKIP (https://nvbugs/5512734)
 full:H100/accuracy/test_llm_api_pytorch.py::TestLlama4MaverickInstruct::test_fp8_eagle3[tp8-torch_compile=True] SKIP (https://nvbugs/5483534)
 full:A100/test_e2e.py::test_ptp_quickstart_multimodal[NVILA-8B-FP16-vila/NVILA-8B-video-False] SKIP (https://nvbugs/5453725)
-test_e2e.py::test_ptp_scaffolding[DeepSeek-R1-Distill-Qwen-7B-DeepSeek-R1/DeepSeek-R1-Distill-Qwen-7B] SKIP (https://nvbugs/5517260)
 test_e2e.py::test_ptp_quickstart_multimodal[NVILA-8B-FP16-vila/NVILA-8B-image-False] SKIP (https://nvbugs/5509024)
 test_e2e.py::test_ptp_quickstart_multimodal[qwen2.5-vl-7b-instruct-Qwen2.5-VL-7B-Instruct-video-False] SKIP (https://nvbugs/5509024)
 test_e2e.py::test_ptp_quickstart_multimodal[qwen2.5-vl-7b-instruct-Qwen2.5-VL-7B-Instruct-video-True] SKIP (https://nvbugs/5509024)
@@ -347,3 +346,6 @@ triton_server/test_triton_llm.py::test_llmapi_backend[1-0-enableDecoupleMode-ten
 cpp/test_e2e.py::test_benchmarks[gpt-80] SKIP (https://nvbugs/5601670)
 disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_bf16_empty_batch[DeepSeek-V3-Lite-bf16] SKIP (https://nvbugs/5601682)
 disaggregated/test_disaggregated.py::test_disaggregated_benchmark_on_diff_backends[llama-v3-8b-hf] SKIP (https://nvbugs/5587574)
+full:H20-3e/accuracy/test_llm_api_pytorch.py::TestNemotronUltra::test_auto_dtype[tp8ep4-cuda_graph=True] SKIP (slow I/O)
+full:H20-3e/accuracy/test_llm_api_pytorch.py::TestKimiK2::test_fp8_blockscale[latency] SKIP (slow I/O)
+full:H20-3e/test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[DeepSeek-V3-671B-FP8-DeepSeek-V3-0324-8] SKIP (slow I/O)