All GPUs
GPU Catalog / NVIDIA

RTX PRO 6000 Blackwell
GPU infrastructure.

A high-memory Blackwell professional GPU for teams that need AI compute and advanced graphics capability in the same system.

Architecture
Blackwell
GPU memory
96GB GDDR7 per GPU
Published from
$2.14 / hr
Configure GPU Discuss workload
Blackwell architecture96GB GDDR7 per GPU
Published pricing

Available configurations.

Current Geodd hourly rates. GPU memory totals reflect the accelerator count in each published configuration.

ConfigurationVariantTotal GPU memoryHourly rateAction
1x GPUgpu_1x_pro_6000_blackwellRTX PRO 6000 Blackwell96GB GDDR796GB GDDR7 each$2.14per hourDeploy
2x GPUgpu_2x_pro_6000_blackwellRTX PRO 6000 Blackwell192GB GDDR796GB GDDR7 each$4.28per hourDeploy
4x GPUgpu_4x_pro_6000_blackwellRTX PRO 6000 Blackwell384GB GDDR796GB GDDR7 each$8.56per hourDeploy
8x GPUgpu_8x_pro_6000_blackwellRTX PRO 6000 Blackwell768GB GDDR796GB GDDR7 each$17.12per hourDeploy
Hardware profile

Where RTX PRO 6000 Blackwell fits.

RTX PRO 6000 Blackwell combines 96GB of GDDR7 memory with NVIDIA Blackwell architecture. It is positioned for AI development and inference, photorealistic rendering, 3D graphics, and scientific computing.

Choose this family when a workload needs more local memory than 48GB professional GPUs provide and also benefits from Blackwell-generation AI and graphics features.

NVIDIA product information
  1. 01

    AI development and inference

    Use the 96GB memory envelope for larger models, evaluation pipelines, and inference work that needs a professional GPU platform.

  2. 02

    3D rendering and visualization

    Support photorealistic rendering, complex scenes, and interactive visualization with one GPU family built for AI and graphics.

  3. 03

    Scientific computing

    Run GPU-accelerated scientific and technical workloads that benefit from a large local memory pool.

Deployment sizing

Size before you provision.

Hardware fit depends on the model, precision, batch size, software stack, and total working set. Published use cases are not performance guarantees.

  1. 01 / Memory

    Confirm the working set

    Include weights, caches, activations, batches, and framework overhead.

  2. 02 / Software

    Validate the stack

    Check CUDA, driver, framework, precision, and container requirements.

  3. 03 / Capacity

    Confirm provisioning

    Published pricing does not guarantee immediate capacity in every configuration.

Compare hardware
Browse all GPUs