Skip to content
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

docs fix: nvidia quotes GB on their GPUs not GiB #2797

Merged
merged 1 commit into from
Jan 23, 2025
Merged

docs fix: nvidia quotes GB on their GPUs not GiB #2797

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
14 changes: 6 additions & 8 deletions modal/gpu.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,7 +28,7 @@ class T4(_GPUConfig):
"""
[NVIDIA T4 Tensor Core](https://www.nvidia.com/en-us/data-center/tesla-t4/) GPU class.

A low-cost data center GPU based on the Turing architecture, providing 16GiB of GPU memory.
A low-cost data center GPU based on the Turing architecture, providing 16GB of GPU memory.
"""

def __init__(
Expand All @@ -45,7 +45,7 @@ class L4(_GPUConfig):
"""
[NVIDIA L4 Tensor Core](https://www.nvidia.com/en-us/data-center/l4/) GPU class.

A mid-tier data center GPU based on the Ada Lovelace architecture, providing 24GiB of GPU memory.
A mid-tier data center GPU based on the Ada Lovelace architecture, providing 24GB of GPU memory.
Includes RTX (ray tracing) support.
"""

Expand All @@ -63,23 +63,21 @@ class A100(_GPUConfig):
"""
[NVIDIA A100 Tensor Core](https://www.nvidia.com/en-us/data-center/a100/) GPU class.

The flagship data center GPU of the Ampere architecture. Available in 40GiB and 80GiB GPU memory configurations.
The flagship data center GPU of the Ampere architecture. Available in 40GB and 80GB GPU memory configurations.
"""

def __init__(
self,
*,
count: int = 1, # Number of GPUs per container. Defaults to 1.
size: Union[str, None] = None, # Select GiB configuration of GPU device: "40GB" or "80GB". Defaults to "40GB".
size: Union[str, None] = None, # Select GB configuration of GPU device: "40GB" or "80GB". Defaults to "40GB".
):
if size == "40GB" or not size:
super().__init__(api_pb2.GPU_TYPE_A100, count, "A100-40GB", 40)
elif size == "80GB":
super().__init__(api_pb2.GPU_TYPE_A100_80GB, count, "A100-80GB", 80)
else:
raise ValueError(
f"size='{size}' is invalid. A100s can only have memory values of 40GB or 80GB."
)
raise ValueError(f"size='{size}' is invalid. A100s can only have memory values of 40GB or 80GB.")

def __repr__(self):
if self.memory == 80:
Expand All @@ -92,7 +90,7 @@ class A10G(_GPUConfig):
"""
[NVIDIA A10G Tensor Core](https://www.nvidia.com/en-us/data-center/products/a10-gpu/) GPU class.

A mid-tier data center GPU based on the Ampere architecture, providing 24 GiB of memory.
A mid-tier data center GPU based on the Ampere architecture, providing 24 GB of memory.
A10G GPUs deliver up to 3.3x better ML training performance, 3x better ML inference performance,
and 3x better graphics performance, in comparison to NVIDIA T4 GPUs.
"""
Expand Down
Loading