The RTX PRO 6000 Blackwell Server Edition is the newest 96 GB card on DeployGPU: GDDR7 memory instead of the GDDR6 or HBM on every other card here, FP4 and FP8 Tensor Core support, and a single-GPU ceiling high enough to hold models that used to need two 48 GB cards. Every offer on this page is a VM you can launch now with SSH access, billed by the hour and settled to the second when you stop.
- Offers now
- 6
- From
- $2.19/GPU-hr
- GPUs per VM
- 1x, 2x, 4x
- Providers
- 3
RTX PRO 6000 offers you can launch now
The table lists every RTX PRO 6000 VM in stock, cheapest first, and refreshes every 15 minutes. The Launch link opens that exact offer on the deploy page, where the full hourly breakdown is shown before anything is charged.
| GPU | GPUs | Provider | Region | vCPU / RAM / Disk | Interconnect | Per GPU-hr | Per hour | Launch |
|---|---|---|---|---|---|---|---|---|
| RTX PRO 6000 Blackwell 96 GB | 1x | Massed Compute | Iowa, US | 16 / 144 GB / 725 GB | - | $2.19 | $2.19 | Launch |
| RTX PRO 6000 Blackwell 96 GB | 1x | Verda | Helsinki, FI | 30 / 90 GB / - | - | $2.30 | $2.30 | Launch |
| RTX PRO 6000 Blackwell 96 GB | 1x | Verda | Helsinki, FI | 30 / 90 GB / - | - | $2.35 | $2.35 | Launch |
| RTX PRO 6000 Blackwell 96 GB | 1x | QuantaCloud | US | 16 / 144 GB / 725 GB | - | $2.39 | $2.39 | Launch |
| RTX PRO 6000 Blackwell 96 GB | 2x | Verda | Helsinki, FI | 60 / 180 GB / - | PCIe | $2.30 | $4.60 | Launch |
| RTX PRO 6000 Blackwell 96 GB | 4x | Verda | Helsinki, FI | 120 / 360 GB / - | PCIe | $2.30 | $9.21 | Launch |
DeployGPU only lists providers it can launch machines on. For prices across the whole market, compare all providers on GpuPerHour.
What is different about this card
| Spec | RTX PRO 6000 Blackwell Server Edition |
|---|---|
| Memory | 96 GB GDDR7 |
| Memory bandwidth | 1,597 GB/s |
| Memory interface | 512-bit |
| GPU-to-GPU link | None. PCIe only |
| Low precision | FP4 and FP8 Tensor Core support |
| Power | Up to 600 W, configurable |
Specifications from NVIDIA's RTX PRO 6000 Blackwell Server Edition page. GDDR7 is new to this generation of NVIDIA cards, and 1,597 GB/s puts this card ahead of every 48 GB workstation GPU on bandwidth, though still well under the HBM3e on an H200 or B200. There is no NVLink, so a multi-GPU RTX PRO 6000 VM talks over PCIe between cards, the same as an L40S or RTX 6000 Ada.
What fits in 96 GB
These are weights-only sizes of the published checkpoints, calculated rather than measured. The space left over is what vLLM has for the KV cache after its default 92% memory limit.
| Model | Format | Weights | On one RTX PRO 6000 96 GB |
|---|---|---|---|
| Qwen3.8-27B | BF16 | 55.6 GB | Fits, about 33 GB left for the KV cache |
| Gemma 4 31B | BF16 | 62.5 GB | Fits, about 26 GB left |
| gpt-oss-120b | MXFP4 | 65.2 GB | Fits, about 23 GB left |
| Qwen3.6-35B-A3B | BF16 | 71.9 GB | Fits, about 16 GB left |
| Llama 3.3 70B | BF16 | 141.1 GB | Does not fit. Use 2x RTX PRO 6000 |
The 96 GB ceiling is the reason this card shows up between the 80 GB and 141 GB tiers: gpt-oss-120b, which leaves only about 8 GB of headroom on an H100, leaves close to three times that here, without paying for an H200.
How long until it is ready
A VM usually reaches Running in 1 to 3 minutes. App templates take about 6 to 8 minutes, because the platform waits until the app itself answers before it marks the deployment ready. On some providers the machine reports Running a little before SSH accepts connections, so a Connection refused in the first minute usually clears on its own. The Connection refused fix covers what to check if it does not.
Paid time starts when provisioning starts, but only a launch that reaches Running is charged. A launch that fails refunds the reserved hour automatically.
One GPU or eight
RTX PRO 6000 VMs come in fixed sizes of 1, 2, 4 or 8 GPUs, and the GPUs column shows what is in stock right now. Because this card has no NVLink, the Interconnect column reads PCIe at every size. One VM is the largest unit DeployGPU launches: there is no self-serve multi-node cluster. For more than 8 RTX PRO 6000s, or a fixed term at a fixed price, request a cluster quote and a person replies by email. The multi-GPU page goes deeper on how GPUs inside one node talk to each other over PCIe.
Software on the machine
In the dashboard, every RTX PRO 6000 VM boots an Ubuntu image with NVIDIA drivers and Docker installed by the provider, and you connect over SSH as the user shown on the deployment. Three app templates set up a ready app on a private HTTPS link instead: JupyterLab with PyTorch, ComfyUI and Open WebUI with Ollama. ComfyUI in particular benefits from this card's memory and GDDR7 bandwidth for image and video workloads. For an OpenAI-compatible server on your own RTX PRO 6000, follow the vLLM guide.
Blackwell cards also need newer software than Hopper or Ampere. The images the dashboard offers for these offers carry a driver recent enough for Blackwell, and a launch with an image whose driver is too old is rejected before anything is charged. Your own stack has to keep up too: PyTorch added Blackwell support in version 2.7, with builds for CUDA 12.8 (release notes), so wheels built for older CUDA versions do not run here. Install a current build with:
pip install -U torch --index-url https://download.pytorch.org/whl/cu128
What you pay
The price on each row is the hourly rate you are billed, including any platform fee. Launching reserves the first hour from your credit balance, and when you stop, the unused seconds of that hour go back to the balance. Credits are bought by card with a 5% + $0.30 processing fee on each top-up. The pricing page has the full breakdown.
Stopping an RTX PRO 6000 deletes its disk, so copy results off first. The guide to stopping a GPU without losing work shows how.
Frequently asked questions
How much does an RTX PRO 6000 cost per hour on DeployGPU?
The table on this page lists the current hourly price of every RTX PRO 6000 VM you can launch, cheapest first, refreshed every 15 minutes. The price shown is the rate you are billed, including any platform fee.
Does the RTX PRO 6000 have NVLink?
No. NVIDIA's datasheet lists no GPU-to-GPU link for this card. A multi-GPU RTX PRO 6000 VM connects its cards over PCIe.
How much memory does the RTX PRO 6000 have?
96 GB of GDDR7, more than an A100 or H100 and less than an H200. See the 96 GB and larger comparison page for how it stacks up against the H200 and B200.
Can I rent a single RTX PRO 6000?
Yes, whenever a 1x offer is in stock. The GPUs column shows how many RTX PRO 6000s come with each VM, and you pay only for the time the VM runs.
How long until I can SSH into an RTX PRO 6000?
Most VMs reach Running in 1 to 3 minutes. On some providers SSH starts accepting connections a little after the machine reports Running, so retry for a minute before assuming something is wrong.
Is the RTX PRO 6000 good for ComfyUI or image generation?
Yes. The 96 GB of memory and GDDR7 bandwidth give it room for large diffusion pipelines, and the ComfyUI template launches on it directly from the dashboard.
Do I need a contract to rent an RTX PRO 6000?
No. Every RTX PRO 6000 here is on demand with no minimum term. You prepay credits, launch, and stop whenever you want.