AMD Instinct MI350X vs NVIDIA L40
AMD Instinct MI350X
NVIDIA L40
|
16384 Shaders
288GB HBM3e
2200MHz
|
18176 Shaders
48GB GDDR6
2490MHz
|
|
Peak AI Performance
18.45 PFLOPS
FP4 Tensor Sparse
|
Peak AI Performance
2.9 POPS
INT4 Tensor Sparse
|
|
FP32
144.18 TFLOPS
|
FP32
90.52 TFLOPS
|
|
FP16
144.18 TFLOPS
|
FP16
90.52 TFLOPS
|
|
Form Factor
OAM Module
-
|
Form Factor
PCIe Card
2.0-Slots
|
|
TDP
1000W
|
TDP
300W
|
|
-
-
-
-
-
|
Power Connectors
-
-
-
1x 16-Pin 12VHPWR
|
|
GB6 OpenCL N/A
0%
|
|
|
GB6 Metal N/A
0%
|
GB6 Metal N/A
0%
|
|
GB6 Vulkan N/A
0%
|
|
GB5 OpenCL N/A
0%
|
GB5 OpenCL N/A
0%
|
|
GB5 CUDA N/A
0%
|
GB5 CUDA N/A
0%
|
|
GB5 Metal N/A
0%
|
GB5 Metal N/A
0%
|
|
GB5 Vulkan N/A
0%
|
GB5 Vulkan N/A
0%
|
|
OCT 2020.1 N/A
0%
|
OCT 2020.1 N/A
0%
|
|
OCT Metal N/A
0%
|
OCT Metal N/A
0%
|
|
Peak AI
Performance
18.45 PFLOPS
FP4 Tensor Sparse
|
Peak AI
Performance
2.9 POPS
INT4 Tensor Sparse
|
|
FP4
9.23 PFLOPS
Tensor
18.45 PFLOPS
Tensor Sparse
|
-
-
-
|
|
FP8
-
4.61 PFLOPS
Tensor (FP16 Accumulate)
9.23 PFLOPS
Tensor (FP16 Accumulate) Sparse
4.61 PFLOPS
Tensor (FP32 Accumulate)
9.23 PFLOPS
Tensor (FP32 Accumulate) Sparse
|
FP8
-
724.13 TFLOPS
Tensor (FP16 Accumulate)
1.45 PFLOPS
Tensor (FP16 Accumulate) Sparse
724.13 TFLOPS
Tensor (FP32 Accumulate)
1.45 PFLOPS
Tensor (FP32 Accumulate) Sparse
|
|
FP16
144.18 TFLOPS
2.31 PFLOPS
Tensor (FP16 Accumulate)
4.61 PFLOPS
Tensor (FP16 Accumulate) Sparse
2.31 PFLOPS
Tensor (FP32 Accumulate)
4.61 PFLOPS
Tensor (FP32 Accumulate) Sparse
|
FP16
90.52 TFLOPS
362.07 TFLOPS
Tensor (FP16 Accumulate)
724.13 TFLOPS
Tensor (FP16 Accumulate) Sparse
362.07 TFLOPS
Tensor (FP32 Accumulate)
724.13 TFLOPS
Tensor (FP32 Accumulate) Sparse
|
|
FP32
144.18 TFLOPS
144.18 TFLOPS
Tensor
288.36 TFLOPS
Tensor Sparse
|
FP32
90.52 TFLOPS
-
-
|
|
FP64
72.09 TFLOPS
72.09 TFLOPS
Tensor
|
FP64
1.41 TFLOPS
-
|
|
BF16
-
2.31 PFLOPS
Tensor
4.61 PFLOPS
Tensor Sparse
|
BF16
90.52 TFLOPS
362.07 TFLOPS
Tensor
724.13 TFLOPS
Tensor Sparse
|
|
-
-
-
|
TF32
181.03 TFLOPS
Tensor
362.07 TFLOPS
Tensor Sparse
|
|
INT4
4.61 POPS
Tensor
9.23 POPS
Tensor Sparse
|
INT4
1.45 POPS
Tensor
2.9 POPS
Tensor Sparse
|
|
INT8
-
4.61 POPS
Tensor
9.23 POPS
Tensor Sparse
|
INT8
-
724.13 TOPS
Tensor
1.45 POPS
Tensor Sparse
|
|
-
-
|
INT32
45.26 TOPS
|
|
-
-
|
Ray Tracing
209.2 TOPS
|
|
Pixel
Fillrate
-
|
Pixel
Fillrate
478.08 GPixel/s
|
|
-
-
|
-
-
|
|
Texture
Fillrate
-
|
Texture
Fillrate
1414.32 GTexel/s
|
|
Manufacturer
AMD
|
Manufacturer
NVIDIA
|
|
Chip Designer
AMD
|
Chip Designer
NVIDIA
|
|
Architecture
CDNA 4
|
Architecture
Ada Lovelace
|
|
Family
Instinct
|
Family
Server
|
|
Codename
CDNA Next
-
-
-
|
Codename
NV182
AD102
Variant
AD102-895-A1
|
|
Market Segment
Server
|
Market Segment
Server
|
|
Release Date
6/12/2025
|
Release Date
10/13/2022
|
|
Foundry
TSMC
TSMC IO Die
|
Foundry
TSMC
-
|
|
Fabrication Node
N3P
N6 IO Die
|
Fabrication Node
4N
-
|
|
Die Size
8x 181 mm²
-
|
Die Size
608 mm²
-
|
|
Transistor Count
8x 17.4 Billion
2x 23 Billion IO Die
|
Transistor Count
76.3 Billion
-
|
|
Transistor Density
96.13M/mm²
-
|
Transistor Density
125.41M/mm²
-
|
|
Form
OAM Module
|
Form
PCIe Card
|
|
Shading Units
16384 Shaders
-
|
Shading Units
18176 Shaders
-
|
|
Texture Mapping Units
-
|
Texture Mapping Units
568 TMUs
|
|
Render Output Units
-
|
Render Output Units
192 ROPs
|
|
Tensor Cores
1024 T-Cores
|
Tensor Cores
568 T-Cores
|
|
-
-
|
Ray-Tracing Cores
142 RT-Cores
|
|
-
-
|
Streaming Multiprocessors
142 SMs
|
|
Compute Units
256 CUs
|
-
-
|
|
-
-
|
-
-
|
|
-
-
|
Graphics Processing Clusters
12 GPCs
|
|
-
-
-
2200MHz
|
-
-
735MHz Base
2490MHz
|
|
-
-
|
-
-
|
|
L1
-
-
16KB/CU
-
|
L1
64KB/SM Tex
128KB/SM
-
-
|
|
L2
16MB Shared
|
L2
96MB Shared
|
|
L3
256MB Shared
-
|
-
-
-
|
|
288GB
HBM3e
ECC
|
48GB
GDDR6
-
|
|
Bus Width
8192Bit
|
Bus Width
384Bit
|
|
Clock
3906MHz
Transfer Rate
7.8GT/s
Bandwidth
8000.1GB/s
|
Clock
2250MHz
Transfer Rate
18GT/s
Bandwidth
864GB/s
|
|
-
-
-
-
-
-
-
-
-
|
-
-
-
-
-
-
-
-
-
|
|
TDP
1000W
|
TDP
300W
|
|
-
-
|
-
-
|
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
No Ports
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
4x DisplayPort 1.4
-
-
-
-
-
-
-
-
-
-
-
-
-
-
|
|
Max Resolution
Unknown
|
Max Resolution
7680x4320
|
|
Max Resolution Refresh Rate
-
|
Max Resolution Refresh Rate
120Hz
|
|
Variable Refresh Rate
-
-
-
|
Variable Refresh Rate
G-Sync
FreeSync
-
|
|
Display Stream Compression (DSC)
Not Supported
|
Display Stream Compression (DSC)
Supported
|
|
Multi Monitor Support
Unknown
|
Multi Monitor Support
4
|
|
-
-
|
Content Protection
HDCP 2.3
|
|
Model
4x VCN 4.0
|
Model
2x NVENC 8
|
|
Codec
-
-
-
-
-
-
VP9
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
|
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
|
|
Model
4x VCN 4.0
|
Model
NVDEC 5
|
|
Codec
MPEG-1
MPEG-2
MPEG-4
JPEG
VC-1
-
VP9
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
|
Codec
MPEG-1
MPEG-2
MPEG-4
-
VC-1
VP8
VP9
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
|
|
-
-
-
-
|
Direct X
12
Direct 3D
12_3
|
|
-
-
OpenCL
3.0
-
-
|
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
|
|
-
-
-
-
GFX
9.4
-
-
-
-
|
Shader Model
6.7
CUDA
8.9
-
-
PureVideo HD
VP12
VDPAU
Feature Set L
|
|
-
-
-
-
|
-
-
-
-
|
|
-
-
-
-
-
-
-
-
|
Power Connectors
-
-
-
-
-
-
1x 16-Pin 12VHPWR
|
|
-
-
PCIe Version
5.0
PCIe Lanes
16
|
Slots Required
2.0
PCIe Version
4.0
PCIe Lanes
16
|
|
Multi GPU Support
Supported
Type
Infinity Fabric
|
-
-
-
-
|
|
-
-
-
-
-
-
|
Height
111 mm (4.37 in)
Width
267 mm (10.51 in)
Depth
40 mm (1.57 in)
|
Copy Link