Rent an NVIDIA RTX 6000D GPU with 84 GB VRAM
Rent RTX 6000D cloud GPU servers by the hour on Cloudj: 84 GB VRAM on NVIDIA Blackwell with about 77.6 TFLOPS FP32 and 1222 GB/s of memory bandwidth. It brings 80 GB-class memory to a workstation card, for AI teams and studios that run 70B-class LLMs, long-context inference and heavy generative pipelines at a 600 W power limit. Use CUDA 12.8 or newer builds of your software, pick a live configuration and connect over SSH with your key.
RTX 6000D specifications
VRAM per GPU
84 GB
Architecture
NVIDIA Blackwell
Compute capability
12
FP32 compute
≈ 77.6 TFLOPS
Memory bandwidth
≈ 1222 GB/s
Power limit
600 W
Values reported by the infrastructure provider for machines currently listed; real performance varies by machine.
Live RTX 6000D configurations and prices
Out of stock right now
Browse available GPUsWhat you can run on the RTX 6000D cloud GPU
70B models with long context
84 GB holds a 4-bit 70B model with plenty of KV cache, or 30B–34B models in 16-bit.
Large video generation workflows
Run memory-heavy video and image models in ComfyUI at full precision without offloading layers.
Fine-tuning 30B–70B models
QLoRA on 70B-class models and LoRA on 30B-class models fit on one Blackwell GPU.
Multi-model AI pipelines
Keep an LLM, an embedding model and an image model loaded side by side for agent or RAG prototypes.
Is the RTX 6000D the right GPU for your workload?
Choose the RTX 6000D if your workloads require its substantial 84 GB VRAM capacity and Blackwell architecture features on a single workstation card. Opt for a smaller consumer or workstation GPU for lightweight inference to lower costs, or step up to a datacenter GPU if your workflow requires higher memory bandwidth.
RTX 6000D vs similar cloud GPUs
Guides for RTX 6000D
RTX 6000D cloud GPU: frequently asked questions
Rent the RTX 6000D cloud GPU by the hour
Pick a live configuration, choose a template and connect over SSH with your key. Billing stops when you destroy the instance.