RDNA 2 has no matrix cores. AMD's 26.62 TFLOPS is packed FP16 vector math at exactly twice the FP32 vector rate. AMD publishes no INT8, INT4, BF16, FP8 or FP64 number for this card. The 1,326 GB/s effective figure is AMD's blended Infinity Cache number and is not comparable with the 432 GB/s that crosses the GDDR6 interface.

AMD Radeon RX 6750 XT 12GB

RDNA 2 PCIe 2022 7nm Spec confidence: Official
FP32
13
TFLOPS
VRAM
12 GB
GDDR6
Bandwidth
432 GB/s
memory
TDP
250 W
53.2 TFLOPS/kW
Download datasheet

Overview

The Radeon RX 6750 XT is the 2022 refresh of the RX 6700 XT: the same 40 compute unit Navi 22 die and the same 12 GB across a 192-bit bus, but with 18 Gbps memory instead of 16 Gbps and a 250 W board power instead of 230 W. Peak FP32 barely moves, from 13.21 to 13.31 TFLOPS, while bandwidth climbs from 384 to 432 GB/s, which tells you where the refresh actually spent its budget. RDNA 2 has no matrix cores, so the 26.62 TFLOPS FP16 AMD quotes is packed vector throughput at twice FP32, and no BF16, FP8, INT8 or INT4 rate is published for the generation at all.

Performance

Peak theoretical throughput by precision type

PrecisionPeak
FP64
No verified data available
FP32
32-bit floating point
13TFLOPS
TF32
No verified data available
BF16
No verified data available
FP16
16-bit floating point
27TFLOPS
FP8
No verified data available
FP6
No verified data available
FP4
No verified data available
INT8
No verified data available

Every figure here is dense. The vendor documents no structured (2:4) sparsity mode for this part, so there is no second number to quote.

The Radeon RX 6750 XT 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 MI355XRadeon RX 6750 XT 12GB

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

Specifications

Architecture

RDNA 2

Form Factor

PCIe

Launch Year

2022

Process Node

7nm

Memory

12 GB GDDR6

Bandwidth

432 GB/s

TDP

250 W

Max Power

~288 W

Transistors

17.2 billion

Stream Processors

2,560

Spec Confidence

Official

Full Specifications

Compute Engine
Stream Processors 2,560
Compute Units 40
Base Clock 2.15 GHz
Boost Clock 2.60 GHz
Chip Design
Transistors 17.2 billion
Process Node 7nm
Chiplets 1 (Monolithic)
Memory
VRAM 12 GB
Memory Type GDDR6
Bandwidth 432 GB/s
Interface Width 192-bit
Effective Bandwidth 1.3 TB/s
Memory Clock 18 GT/s
Cache
Infinity Cache 96 MB
Power & Thermal
TDP 250 W
Max Board Power The vendor publishes no maximum board power; the Max Power tile above estimates it from TDP
Power Connector 1x8-Pin + 1x6-Pin
Enterprise Features
Sparsity No
Compute APIs DirectX 12, OpenGL, OpenCL, Vulkan
Physical & Media
Card Length 267 mm
Width 2-slot
General
Form Factor PCIe
Architecture RDNA 2
Process Node 7nm
Launch Year 2022

Datasheet & Resources

Data Provenance

Every figure traced to a source

Primary Source

Publisher
AMD
Published
No verified data available

Data Quality

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

Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.

Browse GPU Cloud Providers

Similar GPUs

Frequently Asked Questions

How many TFLOPS does the AMD Radeon RX 6750 XT have?

The AMD Radeon RX 6750 XT delivers 13 TFLOPS FP32 and 27 TFLOPS FP16 at peak. Flopper does not currently have a verified FP8 throughput figure for it.

What is the power consumption of the AMD Radeon RX 6750 XT?

The AMD Radeon RX 6750 XT has a TDP (Thermal Design Power) rating of 250 watts.

How much memory does the AMD Radeon RX 6750 XT have?

The AMD Radeon RX 6750 XT is equipped with 12 GB of VRAM with 432 GB/s of memory bandwidth.

What architecture is the AMD Radeon RX 6750 XT based on?

The AMD Radeon RX 6750 XT is based on the RDNA 2 architecture, launched in 2022.

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