geekradar
Int
Explore GeekRadar

More to explore

Good gear All topics About us Say hello
GeekRadar / The comparison desk

COMPARE.

GPUs
Filters

Filter GPUs

Brand
Benchmark coverage
Clear all filters

Best result per entry; configurations may differ.

Type
266 results Benchmark score
NVIDIA

NVIDIA GeForce RTX 3090 Ti

GPU
Video memory
24 GB
Rated memory bandwidth
1,008.30 GB/s
Thermal design power
450 W
GPU cores
10,752*
Geekbench GPU (reported compilation) · version unreported · CUDA#1* · 260,346*Platform unreported
Blender 5.1.1 / Linux / OPTIX5,601.47 samples/minRendering · median · 7 complete runs
Nvidia

A100 80GB PCIe

GPU*
Video memory
80 GB*
Rated memory bandwidth
1,935 GB/s*
Architecture
Ampere*
Video memory type
HBM2e*
Geekbench GPU (reported compilation) · version unreported · CUDA#2* · 259,828*Platform unreported
No Blender rendering score
Nvidia

A100-PCIE-80GB

GPU*
Video memory
80 GB*
Rated memory bandwidth
1,935 GB/s*
Architecture
Ampere*
Video memory type
HBM2e*
Geekbench GPU (reported compilation) · version unreported · CUDA#3* · 256,292*Platform unreported
No Blender rendering score
NVIDIA

NVIDIA GeForce RTX 3090

GPU
Video memory
24 GB
Rated memory bandwidth
936.20 GB/s
Thermal design power
350 W
GPU cores
10,496*
Geekbench GPU (reported compilation) · version unreported · CUDA#4* · 238,123*Platform unreported
Blender 5.1.1 / Windows / OPTIX5,177.84 samples/minRendering · median · 59 complete runs
Nvidia

A100-SXM4-40GB

GPU*
Video memory
40 GB*
Rated memory bandwidth
1,555 GB/s*
Architecture
Ampere*
Video memory type
HBM2*
Geekbench GPU (reported compilation) · version unreported · CUDA#5* · 237,220*Platform unreported
No Blender rendering score
NVIDIA

NVIDIA GeForce RTX 3080 Ti

GPU
Video memory
12 GB
Rated memory bandwidth
912.40 GB/s
Thermal design power
350 W
GPU cores
10,240*
Geekbench GPU (reported compilation) · version unreported · CUDA#6* · 235,513*Platform unreported
Blender 5.1.1 / Windows / OPTIX5,028.76 samples/minRendering · median · 61 complete runs
Nvidia

GRID A100-7-40C MIG 7g.40gb

GPU*
Geekbench GPU (reported compilation) · version unreported · CUDA#7* · 233,910*Platform unreported
No Blender rendering score
Unspecified

RTX A6000

GPU*
Video memory
48 GB*
Rated memory bandwidth
768 GB/s*
Architecture
Ampere*
Process node
8 nm*
Geekbench GPU (reported compilation) · version unreported · CUDA#8* · 224,604*Platform unreported
No Blender rendering score
Nvidia

GRID A100-4C

GPU*
Geekbench GPU (reported compilation) · version unreported · CUDA#9* · 219,037*Platform unreported
No Blender rendering score
NVIDIA

NVIDIA Tesla V100S-PCIE-32GB

GPU
Video memory
32 GB*
Rated memory bandwidth
1,134 GB/s*
Architecture
Volta
Manufacturing process
12 nm
Geekbench GPU (reported compilation) · version unreported · CUDA#10* · 216,224*Platform unreported
Blender 4.1.0 / Linux / CUDA2,847.18 samples/minRendering · median · 3 complete runs
Nvidia

A100-PCIE-40GB

GPU*
Video memory
40 GB*
Rated memory bandwidth
1,555 GB/s*
Architecture
Ampere*
Video memory type
HBM2*
Geekbench GPU (reported compilation) · version unreported · CUDA#11* · 215,434*Platform unreported
No Blender rendering score
NVIDIA

NVIDIA GeForce RTX 3080

GPU
Video memory
10 GB
Rated memory bandwidth
760.30 GB/s
Thermal design power
320 W
GPU cores
8,704*
Geekbench GPU (reported compilation) · version unreported · CUDA#12* · 206,390*Platform unreported
Blender 5.1.1 / Windows / OPTIX4,274.87 samples/minRendering · median · 89 complete runs
Nvidia

A40

GPU*
Video memory
48 GB*
Rated memory bandwidth
696 GB/s*
Architecture
Ampere*
Video memory type
GDDR6*
Geekbench GPU (reported compilation) · version unreported · CUDA#13* · 205,120*Platform unreported
No Blender rendering score
NVIDIA

NVIDIA Tesla V100-SXM2-32GB

GPU
Video memory
32 GB*
Rated memory bandwidth
900 GB/s*
Thermal design power
300 W*
Architecture
Volta
Geekbench GPU (reported compilation) · version unreported · CUDA#14* · 203,550*Platform unreported
Blender 3.4.0 / Linux / OPTIX2,265.41 samples/minRendering · median · 3 complete runs
NVIDIA

NVIDIA Tesla V100-SXM2-16GB

GPU
Video memory
16 GB*
Rated memory bandwidth
900 GB/s*
Thermal design power
300 W*
Architecture
Volta
Geekbench GPU (reported compilation) · version unreported · CUDA#15* · 201,257*Platform unreported
Blender 5.1.1 / Windows / OPTIX1,758.23 samples/minRendering · median · 30 complete runs
Unspecified

RTX A5000

GPU*
Video memory
24 GB*
Rated memory bandwidth
768 GB/s*
Architecture
Ampere*
Process node
8 nm*
Geekbench GPU (reported compilation) · version unreported · CUDA#16* · 193,384*Platform unreported
No Blender rendering score
NVIDIA

NVIDIA Tesla V100-PCIE-16GB

GPU
Video memory
16 GB*
Rated memory bandwidth
900 GB/s*
Thermal design power
250 W*
Architecture
Volta
Geekbench GPU (reported compilation) · version unreported · CUDA#17* · 192,531*Platform unreported
Blender 3.4.0 / Linux / OPTIX2,228.00 samples/minRendering · median · 3 complete runs
NVIDIA

NVIDIA Tesla V100-PCIE-32GB

GPU
Video memory
32 GB*
Rated memory bandwidth
900 GB/s*
Thermal design power
250 W*
Architecture
Volta
Geekbench GPU (reported compilation) · version unreported · CUDA#18* · 191,375*Platform unreported
Blender 3.6.0 / Windows / OPTIX2,252.38 samples/minRendering · median · 3 complete runs
Nvidia

A10G

GPU*
Video memory
24 GB*
Rated memory bandwidth
600.20 GB/s*
Architecture
Ampere*
Process node
8 nm*
Geekbench GPU (reported compilation) · version unreported · CUDA#19* · 187,763*Platform unreported
No Blender rendering score
Nvidia

A10

GPU*
Video memory
24 GB*
Rated memory bandwidth
600 GB/s*
Architecture
Ampere*
Video memory type
GDDR6*
Geekbench GPU (reported compilation) · version unreported · CUDA#20* · 186,108*Platform unreported
No Blender rendering score
Nvidia

A40-8Q

GPU*
Geekbench GPU (reported compilation) · version unreported · CUDA#21* · 183,852*Platform unreported
No Blender rendering score
NVIDIA

NVIDIA Quadro GV100

GPU
Video memory
32 GB
Rated memory bandwidth
870 GB/s
Thermal design power
250 W
GPU cores
5,120*
Geekbench GPU (reported compilation) · version unreported · CUDA#22* · 180,942*Platform unreported
GFXBench 5.0 · 1080p Manhattan Offscreen695.58* fpsWindows / dx / 1920 x 1080 / offscreen · archived median
Nvidia

GRID V100D-32Q

GPU*
Geekbench GPU (reported compilation) · version unreported · CUDA#23* · 178,886*Platform unreported
No Blender rendering score
Unspecified

TITAN V

GPU*
Video memory
12 GB*
Rated memory bandwidth
651.30 GB/s*
Architecture
Volta*
Process node
12 nm*
Geekbench GPU (reported compilation) · version unreported · CUDA#24* · 176,773*Platform unreported
No Blender rendering score

* Provisional data