The precision ladder on this row was corrected on 2026-09-02. It previously carried FP16 = BF16 = TF32 = FP32 = 34.1 with INT8 at 68.2 and no INT4 row at all, which is the shader-rates-in-tensor-columns defect: those are plain shader figures, not tensor ones. NVIDIA's COMPUTEX 2021 announcement supplies the missing anchor, "34 Shader-TeraFLOPS, 67 RT-TeraFLOPS, 273 Tensor-TeraFLOPS", and 273 is exactly 8x 34.1, the GeForce GA10x sparse FP16-with-FP16-accumulate rate. TF32 was the one value that was already right and is unchanged, because GeForce Ampere genuinely runs TF32 at the plain FP32 rate.

NVIDIA GeForce RTX 3080 Ti 12GB

Architecture: AmpereForm factor: PCIeReleased: 2021Process: 8nmSpec confidence: Official
FP32
34
TFLOPS
Memory
12 GB
GDDR6X
Bandwidth
912 GB/s
memory
TDP
350 W
97.4 FP32 dense TFLOPS/kW of TDP

Performance

Peak theoretical throughput by precision type

PrecisionDense2:4 Sparse
FP64
64-bit floating point
0.5TFLOPS Structured sparsity is a tensor-core feature; this vector precision has no sparse form
FP32
32-bit floating point
34TFLOPS Structured sparsity is a tensor-core feature; this vector precision has no sparse form
TF32
TensorFloat-32
34TFLOPS 68TFLOPS
BF16
Brain Float 16
68TFLOPS 136TFLOPS
FP16
16-bit floating point
68TFLOPS 136TFLOPS
FP8
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP6
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP4
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
INT8
8-bit integer
273TOPS 546TOPS
INT4
4-bit integer
546TOPS 1,091TOPS

The GeForce RTX 3080 Ti 12GB in the GPU landscape

Peak FP16 TFLOPS (dense) against TDP, single-GPU parts tracked by Flopper

06001,2001,8002,4003,0000 W250 W500 W750 W1000 W1250 W1500 WInstinct MI355XGeForce RTX 3080 Ti 12GB

Higher and further left is better: more half-precision throughput for less power.

Specifications

Architecture

Ampere

Form Factor

PCIe

Launch Year

2021

Process Node

8nm

Memory

12 GB GDDR6X

Bandwidth

912 GB/s

TDP

350 W

Max power

350 W

Transistors

28.3 billion

CUDA Cores

10,240

Spec Confidence

Official

Full Specifications

Compute Engine
CUDA Cores 10,240
Tensor Cores 320 (3rd Gen)
Streaming Multiprocessors 80
Base Clock 1.37 GHz
Boost Clock 1.67 GHz
Chip Design
Transistors 28.3 billion
Die Size 628 mm²
Process Node 8nm
Memory
Memory 12 GB
Memory Type GDDR6X
Bandwidth 912 GB/s
Interface Width 384-bit
Interconnect & I/O
PCIe 4.0 x16
Power & Thermal
TDP 350 W
Max power 350 W
Cooling Active
Enterprise Features
ECC Memory No
Sparsity Yes
Compute APIs CUDA, DirectX 12, Vulkan, OpenGL
General
Form Factor PCIe
Architecture Ampere
Process Node 8nm
Launch Year 2021

Datasheet & Resources

Data Provenance

Every figure traced to a source

Primary Source

Publisher
NVIDIA
Published
2021-06-03

Data Quality

Spec confidence
Official
Clock basis
Boost
Core precisions with figures
6 of 9
Normalization
All values in TFLOPS

Rent this GPU

$0.12 / GPU-hr

Lowest of 2 live listings across 2 providers

See all rental prices →

Price history and every provider on the pricing page. Some links are affiliate links.

Similar GPUs

NVIDIA A100 PCIe 40GB

PCIe · 2020
FP32: 20 TFLOPS
Compare vs A100

NVIDIA A100 PCIe 80GB

PCIe · 2021
FP32: 20 TFLOPS
Compare vs A100

NVIDIA A100 SXM4 40GB

SXM · 2020
FP32: 20 TFLOPS
Compare vs A100

NVIDIA A100 SXM4 80GB

SXM · 2020
FP32: 20 TFLOPS
Compare vs A100

NVIDIA A800 PCIe 80GB

PCIe · 2022
FP32: 20 TFLOPS
Compare vs A800

Frequently Asked Questions

How many TFLOPS does the NVIDIA GeForce RTX 3080 Ti have?

The NVIDIA GeForce RTX 3080 Ti delivers 34 TFLOPS FP32 and 68 TFLOPS FP16 at peak. Flopper does not currently have a verified FP8 throughput figure for it.

What is the power consumption of the NVIDIA GeForce RTX 3080 Ti?

The NVIDIA GeForce RTX 3080 Ti has a TDP (Thermal Design Power) rating of 350 watts.

How much memory does the NVIDIA GeForce RTX 3080 Ti have?

The NVIDIA GeForce RTX 3080 Ti is equipped with 12 GB of memory with 912 GB/s of memory bandwidth.

What architecture is the NVIDIA GeForce RTX 3080 Ti based on?

The NVIDIA GeForce RTX 3080 Ti is based on the Ampere architecture, launched in 2021.

How much does the NVIDIA GeForce RTX 3080 Ti cost to rent?

The NVIDIA GeForce RTX 3080 Ti rents from $0.12 per GPU-hour across 2 cloud providers, the lowest of 2 live listings Flopper tracks. The pricing page lists every provider and the price history.

When did the NVIDIA GeForce RTX 3080 Ti come out?

The NVIDIA GeForce RTX 3080 Ti launched in 2021 on the Ampere architecture.

Stay Updated on GPU Releases

Get notified when new GPUs are added or specifications are updated.

Loading verification...

No spam, unsubscribe anytime.

Back to GPUs

© 2026 Flopper.io - Compare the GPUs Powering AI