Tesla M40

GPU Code: GM200

12 GB

GDDR5

1112 MHz

Boost Clock

250 W

TDP

28 nm

Process
Share on Social Media
Share Tesla M40 specifications with others

Technical Specifications

Core Specifications
Architecture: Maxwell 2.0
Process Size: 28 nm
Base Clock: 948 MHz
Boost Clock: 1112 MHz
Shader Units: 3072
SMs: 24
TMUs: 192
ROPs: 96
Memory & Performance
Memory Size: 12 GB
Memory Type: GDDR5
Memory Bandwidth: 288.4 GB/s
Bus Interface: PCIe 3.0 x16
TDP: 250 W
Suggested PSU: 600 W
Cache & Die Information
L1 Cache: 48 KB
L2 Cache: 3.0 MB
Die Size: 601 mm²
Transistors: 8.0 billion
Transistor Density: 13.3 million/mm²

API Support

DirectX
12.1
OpenGL
4.6
Vulkan
1.3
Cuda Compute
5.2 sm_52

Physical Specifications

Board Length: 267 mm
Power Connectors: 8-pin EPS

Quick Facts

Release Date: 2015-11-10
Manufacturer: NVIDIA
Generation: Tesla Maxwell(Mxx)

Performance

Pixel Fill Rate 106.8 GPixel/s
Texture Fill Rate 213.5 GTexel/s
FP32 Performance 6.8 TFLOPS

CUDA Toolkit Compatibility

Cuda Compute
5.2 sm_52
Architecture

Maxwell 2.0

Supported CUDA Toolkits
7.5 8.0 9.0 9.1 9.2 10.0 10.1 10.2 11.0
Note: Support dropped in CUDA 11.1

Maxwell 2.0 architecture (GM200/GM204) - Support dropped in CUDA 11.1