Skip to content

fix: add H200 TFLOPS #1543

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 2 commits into
base: main
Choose a base branch
from
+44 −0
Open

fix: add H200 TFLOPS #1543

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
958577a
fix: add H200 TFLOPS
azzhipa Nov 19, 2025
6aead1e
Merge branch 'main' into fix/h200_tflops
clumsy Nov 20, 2025
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
2 changes: 2 additions & 0 deletions nemo_rl/utils/flops_tracker.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -115,6 +115,8 @@ def is_using_tf32() -> bool:
("NVIDIA A100 80GB PCIe", torch.float32): 312 / 2 if is_using_tf32() else 19.5,
("NVIDIA H100 80GB HBM3", torch.bfloat16): 1979 / 2,
("NVIDIA H100 80GB HBM3", torch.float32): 989 / 2 if is_using_tf32() else 67.0,
("NVIDIA H200", torch.bfloat16): 1979 / 2,
("NVIDIA H200", torch.float32): 989 / 2 if is_using_tf32() else 67.0,
("NVIDIA B200", torch.bfloat16): 4500 / 2,
("NVIDIA B200", torch.float32): 2200 / 2 if is_using_tf32() else 80.0,
("NVIDIA B300", torch.bfloat16): 4500 / 2,
Expand Down
42 changes: 42 additions & 0 deletions tests/unit/utils/test_flops_tracker.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,42 @@
# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

import pytest

import torch

from nemo_rl.utils.flops_tracker import get_theoretical_tflops, is_using_tf32


@pytest.mark.parametrize(
"device_name, model_dtype, tflops",
[
("NVIDIA A100 80GB PCIe", torch.bfloat16, 624 / 2),
("NVIDIA A100 80GB PCIe", torch.float32, 312 / 2 if is_using_tf32() else 19.5),
("NVIDIA H100 80GB HBM3", torch.bfloat16, 1979 / 2),
("NVIDIA H100 80GB HBM3", torch.float32, 989 / 2 if is_using_tf32() else 67.0),
("NVIDIA H200", torch.bfloat16, 1979 / 2),
("NVIDIA H200", torch.float32, 989 / 2 if is_using_tf32() else 67.0),
("NVIDIA B200", torch.bfloat16, 4500 / 2),
("NVIDIA B200", torch.float32, 2200 / 2 if is_using_tf32() else 80.0),
("NVIDIA B300", torch.bfloat16, 4500 / 2),
("NVIDIA B300", torch.float32, 2200 / 2 if is_using_tf32() else 80.0),
("NVIDIA GB200", torch.bfloat16, 4900 / 2),
("NVIDIA GB200", torch.float32, 2500 / 2 if is_using_tf32() else 80.0),
("NVIDIA GB300", torch.bfloat16, 4900 / 2),
("NVIDIA GB300", torch.float32, 2500 / 2 if is_using_tf32() else 80.0),
],
)
def test_theoretical_tflops(device_name, model_dtype, tflops):
assert get_theoretical_tflops(device_name, model_dtype) == pytest.approx(tflops)
Loading