Not the same card as the RTX 4070 Ti. The SUPER moves from AD104 to a cut AD103, from 7,680 cores to 8,448, and from 12 GB on a 192-bit bus to 16 GB on a 256-bit bus, at the same 285 W. Two distinct rows, and listings that say only "4070 Ti" are ambiguous.

NVIDIA GeForce RTX 4070 Ti SUPER 16GB

Ada Lovelace PCIe 2024 4nm Spec confidence: Official
FP8 (dense)
176
TFLOPS
FP32
44
TFLOPS
VRAM
16 GB
GDDR6X
TDP
285 W
154.7 TFLOPS/kW

Overview

The GeForce RTX 4070 Ti SUPER was NVIDIA's January 2024 answer to the criticism that the 12 GB RTX 4070 Ti was short on memory. It swaps the AD104 die for a 66 SM cut of AD103, giving 8,448 CUDA cores and 264 fourth-generation Tensor cores, and widens the memory bus to 256-bit while raising capacity to 16 GB of GDDR6X, all inside the same 285 W envelope. For AI that combination, 16 GB and 176.4 TFLOPS of dense FP8, makes it one of the better value used Ada cards for local inference, sitting between the 12 GB RTX 4070 SUPER and the far more expensive RTX 4080. NVIDIA does not publish memory bandwidth for any RTX 40 card, so that field is deliberately empty here rather than filled from a third party.

Performance

Peak theoretical throughput by precision type

PrecisionDense2:4 Sparse
FP64
64-bit floating point
0.7TFLOPS Structured sparsity is a tensor-core feature; this vector precision has no sparse form
FP32
32-bit floating point
44TFLOPS Structured sparsity is a tensor-core feature; this vector precision has no sparse form
TF32
TensorFloat-32
44TFLOPS The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
BF16
Brain Float 16
88TFLOPS The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP16
16-bit floating point
88TFLOPS The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP8
8-bit floating point
176TFLOPS 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
353TOPS The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision

The GeForce RTX 4070 Ti SUPER 16GB 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 4070 Ti SUPER 16GB

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

Specifications

Architecture

Ada Lovelace

Form Factor

PCIe

Launch Year

2024

Process Node

4nm

Memory

16 GB GDDR6X

Bandwidth

No verified data available

TDP

285 W

Max Power

285 W

Transistors

45.9 billion

CUDA Cores

8,448

Spec Confidence

Official

Full Specifications

Compute Engine
CUDA Cores 8,448
Tensor Cores 264 (4th Gen)
Streaming Multiprocessors 66
Base Clock 2.34 GHz
Boost Clock 2.61 GHz
Chip Design
Transistors 45.9 billion
Die Size 379 mm²
Process Node 4nm
Memory
VRAM 16 GB
Memory Type GDDR6X
Bandwidth No verified data available
Interface Width 256-bit
Interconnect & I/O
PCIe 4.0
Power & Thermal
TDP 285 W
Max Board Power 285 W
Cooling Active
Enterprise Features
Sparsity Yes
General
Form Factor PCIe
Architecture Ada Lovelace
Process Node 4nm
Launch Year 2024

Datasheet & Resources

Data Provenance

Every figure traced to a source

Primary Source

Publisher
NVIDIA
Published
2024-01-08

Data Quality

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

Rent this GPU

$0.12 / GPU-hr

Lowest of 1 live listing across 1 provider

See all rental prices →

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

Similar GPUs

NVIDIA L40S 48GB

PCIe · 2023
FP32: 92 TFLOPS
Compare vs L40S

NVIDIA L40 48GB

PCIe · 2023
FP32: 91 TFLOPS
Compare vs L40

Frequently Asked Questions

How many TFLOPS does the NVIDIA GeForce RTX 4070 Ti SUPER have?

The NVIDIA GeForce RTX 4070 Ti SUPER delivers 44 TFLOPS FP32, 88 TFLOPS FP16 and 176 TFLOPS FP8 at peak.

What is the power consumption of the NVIDIA GeForce RTX 4070 Ti SUPER?

The NVIDIA GeForce RTX 4070 Ti SUPER has a TDP (Thermal Design Power) rating of 285 watts.

How much memory does the NVIDIA GeForce RTX 4070 Ti SUPER have?

The NVIDIA GeForce RTX 4070 Ti SUPER is equipped with 16 GB of VRAM. Flopper does not currently have a verified memory bandwidth figure for it.

What architecture is the NVIDIA GeForce RTX 4070 Ti SUPER based on?

The NVIDIA GeForce RTX 4070 Ti SUPER is based on the Ada Lovelace architecture, launched in 2024.

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