Skip to content

[https://nvbugs/5451028][fix] Lower BS for NemotronUltra tests #7212

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
+4 −11
Open

[https://nvbugs/5451028][fix] Lower BS for NemotronUltra tests #7212

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 4 additions & 11 deletions tests/integration/defs/accuracy/test_llm_api_pytorch.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1962,19 +1962,15 @@ class TestNemotronUltra(LlmapiAccuracyTestHarness):
ids=["tp8", "tp8ep4", "tp8ep8"])
def test_auto_dtype(self, cuda_graph, tp_size, pp_size, ep_size):
with LLM(self.MODEL_PATH,
max_batch_size=32,
max_batch_size=8,
tensor_parallel_size=tp_size,
pipeline_parallel_size=pp_size,
moe_expert_parallel_size=ep_size,
cuda_graph_config=CudaGraphConfig()
if cuda_graph else None) as llm:
# Run only one eval as maximal BS is not large
task = MMLU(self.MODEL_NAME)
task.evaluate(llm)
task = GSM8K(self.MODEL_NAME)
task.evaluate(llm)
# task = GPQADiamond(self.MODEL_NAME)
# task.evaluate(llm,
# extra_evaluator_kwargs=dict(apply_chat_template=True))

@skip_pre_hopper
@pytest.mark.skip_less_device(8)
Expand All @@ -1986,20 +1982,17 @@ def test_auto_dtype(self, cuda_graph, tp_size, pp_size, ep_size):
def test_fp8_prequantized(self, cuda_graph, tp_size, pp_size, ep_size):
model_path = f"{llm_models_root()}/nemotron-nas/Llama-3_1-Nemotron-Ultra-253B-v1-FP8"
with LLM(model_path,
max_batch_size=8,
tensor_parallel_size=tp_size,
pipeline_parallel_size=pp_size,
moe_expert_parallel_size=ep_size,
cuda_graph_config=CudaGraphConfig() if cuda_graph else None,
kv_cache_config=KvCacheConfig(
free_gpu_memory_fraction=0.85)) as llm:
assert llm.args.quant_config.quant_algo == QuantAlgo.FP8
task = MMLU(self.MODEL_NAME)
task.evaluate(llm)
# Run only one eval as maximal BS is not large
task = GSM8K(self.MODEL_NAME)
task.evaluate(llm)
task = GPQADiamond(self.MODEL_NAME)
task.evaluate(llm,
extra_evaluator_kwargs=dict(apply_chat_template=True))


class TestNemotronH(LlmapiAccuracyTestHarness):
Expand Down