Unlock the power of next‑gen visual computing with the NVIDIA RTX PRO 6000 Blackwell Max‑Q Workstation Edition 96GB GDDR7 – the ultimate GPU that turns demanding projects into effortless achievements.
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 96GB GDDR7 - Retail
Built on the cutting‑edge NVIDIA Blackwell architecture, this card delivers 24,064 CUDA cores, 5th‑generation Tensor cores and 4th‑generation Ray Tracing cores. You get 110 TFLOPS of FP32 performance, 333 TFLOPS of RT core power, and a staggering 3,511 AI TOPS, so every render, simulation, or AI workload runs at lightning speed.
The massive 96 GB GDDR7 memory with ECC, a 512‑bit interface and 1,792 GB/s bandwidth ensures you never run out of space or bandwidth, even with the most complex scenes or massive datasets. PCIe 5.0 x16 and four DisplayPort 2.1b connectors let you scale to multiple high‑resolution displays without compromise.
Designed for professional studios, the dual‑slot, full‑height card features active cooling and a 300 W power envelope, while Multi‑Instance GPU (MIG) support lets you partition the GPU into up to four independent instances for flexible workload distribution.
Benefits
- Unmatched AI and ray‑tracing performance for real‑time rendering and deep‑learning tasks.
- Massive 96 GB ECC‑protected memory eliminates bottlenecks in large‑scale projects.
- PCIe 5.0 and high‑bandwidth memory deliver ultra‑fast data transfer and low latency.
- Scalable MIG architecture lets you run multiple workloads on a single GPU.
Uses of the product
- Training and inference of large AI models and neural networks.
- High‑resolution 3D rendering and visual effects for film and game production.
- Complex scientific simulations and data‑intensive research computing.
- Multi‑display professional workstations for video editing and CAD.
Features
- 24,064 CUDA cores paired with 5th‑gen Tensor cores delivering 3,511 AI TOPS.
- 4th‑gen Ray Tracing cores providing 333 TFLOPS of RT performance.
- 96 GB GDDR7 memory with ECC, 512‑bit interface and 1,792 GB/s bandwidth.
- PCIe 5.0 x16, dual‑slot active cooling and 300 W power design for reliable operation.
Specifications
| GPU Architecture |
NVIDIA Blackwell |
| CUDA Cores |
24,064 |
| Tensor Cores |
5th Generation |
| Ray Tracing Cores |
4th Generation |
| AI Performance (TOPS) |
3511 AI TOPS |
| Single-Precision (FP32) Performance |
110 TFLOPS |
| RT Core Performance |
333 TFLOPS |
| GPU Memory |
96 GB GDDR7 with ECC |
| Memory Interface Width |
512-bit |
| Memory Bandwidth |
1,792 GB/s |
| System Interface |
PCIe 5.0 x16 |
| Display Connectors |
4 × DisplayPort 2.1b |
| Max Simultaneous Displays |
>4 × 4096×2160 @ 120 Hz
>4 × 5120×2880 @ 60 Hz
>2 × 7680×4320 @ 60 Hz
|
| Video Engines |
>4 × NVENC (9th Gen)
>4 × NVDEC (6th Gen)
|
| MIG (Multi-Instance GPU) Options |
Up to 4 × 24 GB
Up to 2 × 48 GB
Up to 1 × 96 GB
|
| Total Board Power |
300 W |
| Power Connector |
1 × PCIe CEM5 16-pin |
| Cooling / Thermal Solution |
Active cooling |
| Form Factor |
Dual-slot, full height4.4 in × 10.5 in (length) |
| Graphics APIs |
DirectX 12, Shader Model 6.6, OpenGL 4.6, Vulkan 1.3 |
| Compute APIs |
CUDA 12.8, OpenCL 3.0, DirectCompute |
