Skip to content

NVIDIA · Ada Lovelace · 2022

NVIDIA RTX 4080. Not in stock right now.

16 GB of GDDR6X and 97.5 TFLOPS FP16 non-tensor, for training and inference. The RTX 4090 is on demand now from $0.58/hr per GPU on an 8-GPU machine.

Not in stock right now

RTX 4080 is not in stock right now.

Leave your work email to register interest. These requests decide which GPUs we bring on demand next.

The card

16 GB of GDDR6X. At 736 GB/s.

Memory
16 GB
GDDR6X
Bandwidth
736 GB/s
Memory to compute
FP16
97.5
non-tensor TFLOPS
Per node
2 GPUs
PCIe, 320 W each

Spec sheet

How the RTX 4080 compares. Against its closest alternatives.

FP16 non-tensor

NVIDIA RTX 408097.5 TFLOPS
NVIDIA RTX 4070 Ti80.2 TFLOPS
NVIDIA RTX 309071 TFLOPS

Memory

NVIDIA RTX 408016 GB
NVIDIA RTX 4070 Ti12 GB
NVIDIA RTX 309024 GB
NVIDIA V10032 GB

Bandwidth

NVIDIA RTX 4080736 GB/s
NVIDIA RTX 4070 Ti504 GB/s
NVIDIA RTX 3090936 GB/s
NVIDIA V100900 GB/s

Alternatives

Alternatives to the RTX 4080. Same account, different price and performance.

GPUMemory · FP16 non-tensorStands outPrice gapFrom
NVIDIA RTX 4080This GPU16 GB GDDR6X · 97.5 TFLOPSNot in stock right now
NVIDIA RTX 4070 Ti12 GB GDDR6X · 80.2 TFLOPS-4 GB memoryNot in stock right now
NVIDIA RTX 309024 GB GDDR6X · 71 TFLOPS+8 GB memory$0.41/hr
NVIDIA V10032 GB HBM2+16 GB memory$3.35/hr
NVIDIA A4048 GB GDDR6+32 GB memory$2.66/hr

Full specs

NVIDIA RTX 4080 specifications.

Memory

Memory
16 GB
Type
GDDR6X
Bandwidth
736 GB/s

Compute

FP32
48.7 TFLOPS
FP16 non-tensor
97.5 TFLOPS
INT8
390 TOPS
CUDA cores
9,728

Power and form

TDP
320 W
Form factor
Consumer
Architecture
Ada Lovelace
Released
2022

Node

GPUs per node
2
Interconnect
PCIe
Maker
NVIDIA

Questions

Renting the RTX 4080. What buyers ask first.

Other GPUs

Browse every GPU by architecture.