NVIDIA
NVIDIA RTX PRO 6000
96GB GDDR7 Blackwell-architecture GPU — high-density inference and AIGC acceleration
Reference GPU-hour unit rate for full-server rental
$1.95/GPU-hour
Reference rate only. Actual configuration, billing method and total price are set in the customer contract.
A high-memory bare-metal platform built around NVIDIA RTX PRO 6000 Blackwell GPUs, with 96GB GDDR7 per GPU for inference, content generation and model adaptation.
Illustrative whole-server configuration. Component combinations vary by GPU edition, server platform and available deployment capacity; manufacturer GPU figures are not a performance commitment for a TATC deployment.
Memory
96GB GDDR7
1.8 TB/s bandwidth
Architecture
Blackwell
Family · NVIDIA RTX PRO
AI compute
4000 AI TOPS
FP32: 125 TFLOPS
Power
600W
Manufacturer specification
Highlights
- 96GB GDDR7 with 1.8 TB/s memory bandwidth
- 4000 AI TOPS and 125 TFLOPS FP32
- 600W TDP for high-memory AI, rendering and simulation workloads
Software stack
CUDA 12.x
PyTorch 2.x
Triton
vLLM
Best-fit scenarios
- LLM inference (70B-class models fully in VRAM)
- AIGC image / video generation
- Small-model fine-tuning
Hardware gallery
NVIDIA RTX PRO 6000 and server system
RTX PRO 6000 product and multi-GPU server views show the bare-metal platform components used for high-memory AI workloads.
Explore NVIDIA RTX PRO 6000 Bare Metal
Share your workload profile to compare GPU memory, server capacity, networking and software-stack options.