
An intriguing entry in a recent NBD shipping log has unveiled the new NVIDIA Blackwell GPU, identified as the RTX PRO 6000 X, which boasts an impressive 96 GB of GDDR7 memory.
Introducing the RTX PRO 6000 X: NVIDIA’s Latest GPU for AI and High-Performance Tasks
NVIDIA continues to expand its graphics card lineup, particularly the Blackwell RTX PRO series, building on the legacy of the flagship RTX GeForce 50 series. While the top-tier option for gaming remains the RTX 5090, the professional segment is poised for exciting advancements with the new Blackwell architecture.
A recent shipping manifest revealed the presence of a 96 GB variant of an RTX 6000 Blackwell GPU, earlier hinted at with a singularly referenced RTX 4090 version. This specific model’s complete identity was later confirmed in the shipping documentation.

According to leaker @harukaze5719, the “RTX PRO 6000 X Blackwell”was noted in the shipping logs destined for testing in India. Reports suggest at least two different shipping entities were involved, hinting at a broader distribution of these powerful new graphics cards.
The shipping information indicates that the RTX PRO 6000 Blackwell is equipped with the GB202-870
GPU and a substantial VRAM of 96 GB, setting the stage for demanding workloads, particularly in fields like artificial intelligence.

This striking model features GDDR7 memory on a 512-bit memory bus, also utilized by the GeForce RTX 5090, albeit with a lower 32 GB capacity. The additional memory is particularly advantageous for various intensive applications, allowing the RTX PRO 6000 X to excel in environments demanding high-performance graphics.
Interestingly, the total board power (TBP) of the RTX PRO 6000 X hovers near the RTX 5090, currently estimated to be about 25W higher, a minor increase that could help maximize the card’s capabilities. Notably, the RTX PRO series is expected to include not just this 96 GB powerhouse but potentially another model with a reduced memory capacity.

The second model is speculated to feature 48 GB of GDDR7 memory on a 384-bit interface, a configuration that hasn’t been associated with previous Blackwell GPUs. The RTX 5090 employs a GB202 variant, while the RTX 5080 is linked to the slower GB203, which supports a 256-bit memory bus. Current speculations suggest that this new GPU is also tailored for AI-centric workloads, but details remain scarce.
NVIDIA’s Comprehensive Workstation Graphics Card Portfolio
Graphics Card | RTX PRO 6000 X | RTX 6000 Is Here | RTX A6000 | Quadro RTX 8000 | Quadro RTX 6000 | GV100 Frame |
---|---|---|---|---|---|---|
GPU | Blackwell GPU | Ada Lovelace GPU | Ampere GPU | Turing GPU | Turing GPU | GPU Vault |
GPU WeU | GB202 | AD102 | GA102 | TU102 | TU102 | GV100 |
GPU Process | 5nm | 5nm | 8nm | 12nm | 12nm | 12nm |
Die Size | 750mm² | 608mm² | 628mm² | 754mm² | 754mm² | 815mm² |
GPU Cores | TBD | 18176 Cores | 10752 Cores | 4608 Cores | 4608 Cores | 5120 Cores |
Tensor Cores | TBD | 568 Cores | 656 Cores | 576 Cores | 576 Cores | 640 Cores |
Boost Clock | TBD | 2.50 GHz | 1.80 GHz | 1.77 GHz | 1.77 GHz | 1.62 GHz |
Single Precision | TBD | 91.1 TFLOPs | 38.7 TFLOPs | 16.31 TFLOPs | 16.31 TFLOPs | 16.66 TFLOPs |
Ray Tracing Spec | TBD | 210.6 TFLOPs | 75.4 TFLOPs | 10 GigaRays/Sec | 10 GigaRays/Sec | N/A |
VRAM | 96 GB GDDR7? | 48GB GDDR6 | 48GB GDDR6 | 48GB GDDR6 | 24GB GDDR6 | 32 GB HBM2 |
NVLINK VRAM | N/A | N/A | 96 GB With NVLINK | 96 GB With NVLINK | 48 GB With NVLINK | N/A |
Memory Bus | TBD | 384-bit | 384-bit | 384-bit | 384-bit | 4096-bit |
Memory Bandwidth | TBD | 960GB/s | 768GB/s | 672 GB/s | 672 GB/s | 870GB/s |
TDP | 600W? | 300W | 300W | ~225W | ~200W | 250W |
Launch Price | TBD | $6800 US | $4650 US | $10000 US | $6300 US | $9000 US |
Launch Date | 2025 | Q1 2023 | Q4 2020 | Q4 2018 | Q4 2018 | 2018 |
For more insights and visuals, visit the source.
Leave a Reply ▼