AMD later shipped a 48 GB W7800 on the same 70 compute unit bin, which it names 'AMD Radeon PRO W7800 48GB' and gives its own product page. That card has a 384-bit bus at 864 GB/s and 96 MB of Infinity Cache against this card's 256-bit bus, 576 GB/s and 64 MB, with identical compute. It is a separate row in this catalog under variant 48GB.

AMD Radeon PRO W7800 32GB

RDNA 3 PCIe 2023 5nm Spec confidence: Official
FP32
45
TFLOPS
VRAM
32 GB
GDDR6
Bandwidth
576 GB/s
memory
TDP
260 W
173.8 TFLOPS/kW
Download datasheet

Overview

The Radeon PRO W7800 is the 70 compute unit Navi 31 workstation card, delivering 45.2 TFLOPS of FP32 vector and 90.4 TFLOPS of FP16 matrix through 140 AI accelerators. This original version pairs it with 32 GB of ECC GDDR6 on a 256-bit bus at 576 GB/s and 64 MB of Infinity Cache, in a 260 W double-slot board. It is the card to look at if you want ECC and professional driver certification without the W7900's price or its triple-slot cooler, and AMD followed it with a 48 GB version on the same bin in 2024. As with all RDNA 3, there is no FP8 path, no structured sparsity, and no published FP64 or BF16 rate.

Performance

Peak theoretical throughput by precision type

PrecisionPeak
FP64
No verified data available
FP32
32-bit floating point
45TFLOPS
TF32
No verified data available
BF16
No verified data available
FP16
16-bit floating point
90TFLOPS
FP8
No verified data available
FP6
No verified data available
FP4
No verified data available
INT8
8-bit integer
90TOPS
INT4
4-bit integer
181TOPS

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 PRO W7800 32GB 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 PRO W7800 32GB

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

Specifications

Architecture

RDNA 3

Form Factor

PCIe

Launch Year

2023

Process Node

5nm

Memory

32 GB GDDR6

Bandwidth

576 GB/s

TDP

260 W

Max Power

260 W

Transistors

57.7 billion

Stream Processors

4,480

Spec Confidence

Official

Full Specifications

Compute Engine
Stream Processors 4,480
Tensor Cores 140 (RDNA 3)
Compute Units 70
Chip Design
Transistors 57.7 billion
Process Node 5nm
Memory
VRAM 32 GB
Memory Type GDDR6
Bandwidth 576 GB/s
Interface Width 256-bit
Interconnect & I/O
PCIe Gen 4 x16
Cache
Infinity Cache 64 MB
Power & Thermal
TDP 260 W
Max Board Power 260 W
Power Connector 2x8-Pin
Cooling Active
Enterprise Features
ECC Memory Yes
Sparsity No
Compute APIs ROCm, OpenCL, HIP
Physical & Media
Card Length 280 mm
Width 2-slot
General
Form Factor PCIe
Architecture RDNA 3
Process Node 5nm
Launch Year 2023

Datasheet & Resources

Data Provenance

Every figure traced to a source

Primary Source

Publisher
AMD
Published
2023-04-13

Data Quality

Spec confidence
Official
Clock basis
Boost
Core precisions with figures
3 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 PRO W7800 have?

The AMD Radeon PRO W7800 delivers 45 TFLOPS FP32 and 90 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 PRO W7800?

The AMD Radeon PRO W7800 has a TDP (Thermal Design Power) rating of 260 watts.

How much memory does the AMD Radeon PRO W7800 have?

The AMD Radeon PRO W7800 is equipped with 32 GB of VRAM with 576 GB/s of memory bandwidth.

What architecture is the AMD Radeon PRO W7800 based on?

The AMD Radeon PRO W7800 is based on the RDNA 3 architecture, launched in 2023.

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