RDNA 2 has no matrix cores. AMD's 32.33 TFLOPS is packed FP16 vector math at twice the FP32 vector rate, not a matrix or tensor rate. AMD publishes no INT8, INT4, BF16, FP8 or FP64 number for this card. Same 3,840 stream processors and same 60 compute units as the Radeon PRO W6800, which is clocked higher and reaches 17.8 TFLOPS FP32; the two are the same silicon in different bins.

AMD Radeon RX 6800 16GB

RDNA 2 PCIe 2020 7nm Spec confidence: Official
FP32
16
TFLOPS
VRAM
16 GB
GDDR6
Bandwidth
512 GB/s
memory
TDP
250 W
64.7 TFLOPS/kW
Download datasheet

Overview

The Radeon RX 6800 is the 60 compute unit Navi 21 part, the cheapest way into AMD's 16 GB RDNA 2 tier and the lowest power of the Navi 21 trio at 250 W. It keeps the full 128 MB Infinity Cache and the full 256-bit bus at 512 GB/s, giving up only shader count against the RX 6800 XT. It shares its compute unit count with the workstation Radeon PRO W6800, which runs the same die faster and doubles the memory to 32 GB with ECC. RDNA 2 has no matrix cores, so 16.17 TFLOPS FP32 and a packed 32.33 TFLOPS FP16 are the entire published ladder, with no BF16, FP8, INT8 or INT4 rate anywhere in the generation.

Performance

Peak theoretical throughput by precision type

PrecisionPeak
FP64
No verified data available
FP32
32-bit floating point
16TFLOPS
TF32
No verified data available
BF16
No verified data available
FP16
16-bit floating point
32TFLOPS
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 6800 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 MI355XRadeon RX 6800 16GB

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

Specifications

Architecture

RDNA 2

Form Factor

PCIe

Launch Year

2020

Process Node

7nm

Memory

16 GB GDDR6

Bandwidth

512 GB/s

TDP

250 W

Max Power

~288 W

Transistors

26.8 billion

Stream Processors

3,840

Spec Confidence

Official

Full Specifications

Compute Engine
Stream Processors 3,840
Compute Units 60
Boost Clock 2.10 GHz
Chip Design
Transistors 26.8 billion
Process Node 7nm
Chiplets 1 (Monolithic)
Memory
VRAM 16 GB
Memory Type GDDR6
Bandwidth 512 GB/s
Interface Width 256-bit
Memory Clock 16 GT/s
Cache
Infinity Cache 128 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 2x8-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 2020

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 6800 have?

The AMD Radeon RX 6800 delivers 16 TFLOPS FP32 and 32 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 6800?

The AMD Radeon RX 6800 has a TDP (Thermal Design Power) rating of 250 watts.

How much memory does the AMD Radeon RX 6800 have?

The AMD Radeon RX 6800 is equipped with 16 GB of VRAM with 512 GB/s of memory bandwidth.

What architecture is the AMD Radeon RX 6800 based on?

The AMD Radeon RX 6800 is based on the RDNA 2 architecture, launched in 2020.

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