AMD Radeon PRO V710 28GB
Overview
The Radeon PRO V710 is the one AMD graphics card outside Instinct that most people can actually rent: it is the GPU behind Microsoft Azure's NV v710 v5 virtual machine sizes, offered in fractions from a sixth of a card up to the whole thing. It is a passively cooled single-slot 158 W board with 54 compute units, 28 GB of ECC GDDR6 on an unusual 224-bit bus at 448 GB/s, and 54 MB of Infinity Cache, with AMD MxGPU hardware virtualisation for partitioning. Compute is 27.65 TFLOPS of FP32 vector, 55.3 TOPS of INT8 matrix and 110.59 TOPS of INT4. Note that AMD publishes no FP16 matrix figure for this card at all, which is why its half-precision entry here is the vector rate and reads lower than its Radeon PRO siblings.
Performance
Peak theoretical throughput by precision type
| Precision | Peak |
|---|---|
FP64 | No verified data available |
FP32 32-bit floating point | 28TFLOPS |
TF32 | No verified data available |
BF16 | No verified data available |
FP16 16-bit floating point | 28TFLOPS |
FP8 | No verified data available |
FP6 | No verified data available |
FP4 | No verified data available |
INT8 8-bit integer | 55TOPS |
INT4 4-bit integer | 111TOPS |
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.
Per watt divides the dense figure by the 158 W board power. Where only a sparse figure exists, the cell says so.
The Radeon PRO V710 28GB in the GPU landscape
Peak FP16 TFLOPS (dense) against TDP, single-GPU parts tracked by Flopper
Higher and further left is better: more half-precision throughput for less power.
Specifications
Architecture
RDNA 3
Form Factor
PCIe
Launch Year
2024
Process Node
5nm
Memory
28 GB GDDR6
Bandwidth
448 GB/s
TDP
158 W
Max Power
158 W
Transistors
28.1 billion
Stream Processors
3,456
Spec Confidence
Official
Full Specifications
| Compute Engine | |
|---|---|
| Stream Processors | 3,456 |
| Compute Units | 54 |
| Boost Clock | 2.00 GHz |
| Chip Design | |
| Transistors | 28.1 billion |
| Process Node | 5nm |
| Memory | |
| VRAM | 28 GB |
| Memory Type | GDDR6 |
| Bandwidth | 448 GB/s |
| Interface Width | 224-bit |
| Interconnect & I/O | |
| PCIe | Gen 4 x16 |
| Cache | |
| Infinity Cache | 54 MB |
| Power & Thermal | |
| TDP | 158 W |
| Max Board Power | 158 W |
| Power Connector | 1x8-Pin |
| Cooling | Passive |
| Enterprise Features | |
| ECC Memory | Yes |
| SR-IOV | Yes |
| Sparsity | No |
| Compute APIs | ROCm, OpenCL, HIP |
| Physical & Media | |
| Card Length | 267 mm |
| Width | 1-slot |
| General | |
| Form Factor | PCIe |
| Architecture | RDNA 3 |
| Process Node | 5nm |
| Launch Year | 2024 |
Datasheet & Resources
Flopper Datasheet
AMD Radeon PRO V710 28GB specifications, generated from our database. Printable.
Vendor product page
AMD documentation for the Radeon PRO V710 28GB
AMD Radeon PRO V710 Product Specifications
AMD · 2024-10-03
AMD Instinct Documentation
ROCm guides, architecture details
Data Provenance
Every figure traced to a source
Primary Source
- Publisher
- AMD
- Published
- 2024-10-03
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 ProvidersSimilar GPUs
AMD Radeon PRO W7900 Dual Slot 48GB
Frequently Asked Questions
How many TFLOPS does the AMD Radeon PRO V710 have?
The AMD Radeon PRO V710 delivers 28 TFLOPS FP32 and 28 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 V710?
The AMD Radeon PRO V710 has a TDP (Thermal Design Power) rating of 158 watts.
How much memory does the AMD Radeon PRO V710 have?
The AMD Radeon PRO V710 is equipped with 28 GB of VRAM with 448 GB/s of memory bandwidth.
What architecture is the AMD Radeon PRO V710 based on?
The AMD Radeon PRO V710 is based on the RDNA 3 architecture, launched in 2024.
Stay Updated on GPU Releases
Get notified when new GPUs are added or specifications are updated.
No spam, unsubscribe anytime.