





PNY NVIDIA RTX PRO 6000 Blackwell Workstation Edition
- Free shipping within the EU from €60
- Discreet and privacy-friendly ordering
- Secure payment: iDEAL, Crypto, Klarna, PayPal
- Customers rate us 9.5 - WebwinkelKeur

Pros and cons
Description
PNY NVIDIA RTX PRO 6000 Blackwell: 96 GB of GPU memory for AI and heavy workloads
The NVIDIA RTX PRO 6000 Blackwell Workstation Edition is a professional graphics card for anyone who wants to run large AI models locally or works on demanding creative and technical projects. The card is built on NVIDIA's Blackwell architecture and has 96 GB of GDDR7 memory with ECC. This lets you load models and datasets into memory that do not fit on regular cards.
Computing power for AI
With 24,064 CUDA cores, 752 fifth-generation Tensor Cores and AI performance of 4000 AI TOPS, this GPU is made for inference, fine-tuning and training AI models. It also has more than enough power for simulations, 3D rendering and video editing.
- Single precision: 125 TFLOPS
- RT Core performance: 380 TFLOPS with 188 RT Cores
- 4x NVENC, 4x NVDEC and 4x JPEG decoder for video tasks
Fast and reliable memory
The memory runs over a 512-bit interface with a bandwidth of 1792 GB/s. Thanks to ECC (error correction), your calculations stay reliable even during long sessions. With Multi-Instance GPU (MIG), you can split the card into up to four 24 GB instances or two 48 GB instances. Handy when multiple users or processes need to use the GPU at the same time.
Connections and size
The card uses a PCI Express 5.0 x16 slot and has four DisplayPort 2.1 outputs. It is powered via a single 16-pin PCIe CEM5 connector. Maximum power consumption is 600 W, so make sure your workstation has a robust power supply.
- Size: 5.4" high x 12" long, XHFL dual slot
- Cooling: active, Double Flow Through design
- Frame Lock compatible (with NVIDIA RTX PRO Sync)
- Support for NVIDIA 3D Vision and 3D Vision Pro via 3-pin Mini DIN
- Weight: 1.95 kg
Technical basis
The GPU is manufactured by TSMC on the 4N process developed specifically for NVIDIA. The chip has 92.2 billion transistors on an area of 750 mm².
Want to work locally with the largest AI models? The RTX PRO 6000 Blackwell turns your workstation into a fully-fledged AI machine.
NVIDIA Blackwell
4000 AI TOPS (FP4 with sparsity)
96 GB GDDR7 with ECC (512-bit)
1792 GB/s
600 W (total board power)
Active (double flow-through)
5.4" (H) x 12" (L), dual slot
VCNRTXPRO6000-SB

