AMD earlier today revealed the availability date of its new Radeon AI PRO R9700 graphics card. If you recall, the company had initially unveiled this workstation GPU back in May where it touted Nvidia RTX 5080-besting performance in the case of AI.
The card was supposed to come out in July though that got delayed and now we finally have a date where you will be able to get the card on store shelves.
With this release AMD is promising to bring data-center-class performance into local workstations, and is mainly aimed at developers, researchers, and creators who need to train and run large artificial intelligence (AI) models be it Windows or Linux.
AI developers, your next GPU is here. The Radeon™ AI PRO R9700 brings scalable multi-GPU performance and 32 GB of VRAM to local AI development - available from select retail partners October 27 from $1,299. pic.twitter.com/pzXxZSjJWD
— David McAfee (@McAfeeDavid_AMD) October 23, 2025
Similar to its gaming counterparts, the R9700 is powered by the company’s latest RDNA 4 architecture. AMD highlights the 32GB GDDR6 VRAM on the unit as one of the key specifications and that makes sense given that the R9700 is built for AI and ML number crunching and such workloads are VRAM-demanding.
| Specification | Value |
|---|---|
| Architecture | AMD RDNA 4 |
| GPU memory | 32GB GDDR6 |
| Ray Accelerators | 64 |
| AI Accelerators | 128 |
| Stream Processors | 4096 |
| Compute Units | 64 |
| ROPs | 128 |
| Boost Frequency | Up to 2920 MHz |
| Game Frequency | 2350 MHz |
| Peak Pixel Fill-Rate | Up to 373.76 GP/s |
| Peak Single Precision (FP32 Vector) Performance | 47.8 TFLOPs |
| Peak Half Precision (FP16 Vector) Performance | 95.7 TFLOPs |
| Peak Half Precision (FP16 Matrix) Performance | 191 TFLOPs |
| Peak Half Precision (FP16 Matrix) Performance with Structured Sparsity | 383 TFLOPs |
| Peak 8-bit Precision (FP8 Matrix) Performance (E5M2, E4M3) | 383 TFLOPs |
| Peak 8-bit Precision (FP8 Matrix) Performance with Structured Sparsity (E5M2, E4M3) | 766 TFLOPs |
| Peak 8-bit Precision (INT8 Matrix) Performance | 383 TOPs |
| Peak 8-bit Precision (INT8 Matrix) Performance with Structured Sparsity | 766 TOPs |
| Peak 4-bit Precision (INT4 Matrix) Performance | 766 TOPs |
| Peak 4-bit Precision (INT4 Matrix) Performance with Structured Sparsity | 1531 TOPs |
| Interface | PCI Express Gen 5 |
| Display outputs | Up to 4× DisplayPort 2.1 |
| Video codec support | H.264 encode/decode; H.265/HEVC encode/decode |
| Multi-GPU | Supported via ROCm |
The Radeon AI PRO R9700 begins shipping October 27 for $1299, with broader availability expected through Q4 2025.