China-market export variant of the RTX PRO 6000 for local AI and visualization.
Weekly median price per GPU per hour · Get the data
By provider shows one card per company, placed where its best offer ranked. By configuration lists every offer.
What you can rent comes first: in stock, then waitlist, then not reported, then out of stock. Priced offers rank ahead of quote-only, and on-demand ahead of other billing types.
Within a group, five factors set the order:
We match the provider, its country, GPU model, form factor, billing type, availability and instance name. Matching is partial and case-insensitive. Everyday words work too, so "interruptible" finds spot and "sold out" finds out of stock.
No providers match .
Heads up: A provider's own page may quote a different figure, for example on tax, region or a promotion. A monthly-only plan shows a derived hourly rate. Verify before provisioning. More on how we price.
One RTX PRO 6000D has 84 GB of VRAM. In practice, that's enough memory for roughly 127B parameters at 4-bit or 33B at 16-bit, assuming a 32K context. Below are some open-weight LLMs, with the estimated memory and GPUs each one needs.
| Model | Memory (INT4 / FP4) | RTX PRO 6000Ds needed | Cost /hr | Cost /mo |
|---|---|---|---|---|
Qwen3.8-27B
27B
|
18 GB
|
1
|
–
|
–
|
Gemma 4 31B
30.7B
|
22 GB
|
1
|
–
|
–
|
GPT-OSS-120B
117B · 5.1B active
|
67 GB
|
1
|
–
|
–
|
GLM-5.3-Flash
320B · 18B active
|
178 GB
|
3
|
–
|
–
|
MiniMax-M3
428B · 23B active
|
239 GB
|
4
|
–
|
–
|
DeepSeek V4.1 Flash
552B · 16B active
|
306 GB
|
5
|
–
|
–
|
Kimi K3
2.8T · 104B active
|
1,544 GB
|
21
|
–
|
–
|
Estimates based on the median on-demand rate. Memory is weights plus FP8 KV cache at 32K context per request (FP16/BF16 in the 16-bit column). GPU counts assume 90% of advertised VRAM is usable. No guarantee of runtime support, usable performance, or that a matching quantized build exists. Pricing methodology.
Nvidia RTX PRO 6000D · Per GPU
| Compute · dense | |
|---|---|
| FP4 | 1,553 TFLOPS 3,105 with sparsity |
| FP8 | 776 TFLOPS 1,553 with sparsity |
| FP16 / BF16 | 388 TFLOPS 776 with sparsity |
| INT8 | 776 TOPS 1,553 with sparsity |
| FP32 | 97 TFLOPS |
| Precision support | FP4FP8FP16BF16TF32FP32INT8 |
| Memory | |
|---|---|
| Capacity | 84 GB GDDR7 |
| Bandwidth | 1,568 GB/s |
| Bus width | 448-bit |
| ECC | Yes |
| Silicon | |
|---|---|
| Architecture | Blackwell |
| Process | TSMC 4N |
| Transistors | 92.2 billion |
| Shader cores | 19,968 CUDA cores |
| Matrix cores | 624 Tensor cores |
| Compute units | 156 SMs |
| Fabric and host | |
|---|---|
| Host interface | PCIe 5.0 |
| Power | |
|---|---|
| Board power | 400 W |
As of October 1, 2026, we track 18 configs from 2 providers. Prices are per GPU per hour.
| Billing type | Configs | Cheapest |
|---|---|---|
On-demand | 8 | $0.97 (in stock, GPUhub) |
Reserved | 6 | $1.20 (6 mo, in stock, Vast.ai) |
Spot | 4 | $0.54 (in stock, Vast.ai) |
No median for on-demand, reserved and spot: we only show this when at least 3 providers list the GPU on that billing type.
The cheapest verified in-stock estimate is $698 per month on-demand, $864 reserved (6 mo), $389 spot.
GetDeploying currently tracks RTX PRO 6000D configs from 2 providers. GPUhub and Vast.ai have verified in-stock on-demand configs. See the full price comparison above for every provider and config.
One RTX PRO 6000D runs models up to roughly 127B parameters at 4-bit quantization or 33B at 16-bit, assuming a 32K context. Larger models run across multiple GPUs: the model table above shows the estimated memory and GPU count for popular open-weight LLMs.
84 GB GDDR7 Blackwell GPU sold in the Chinese market, where the full RTX PRO 6000 is export-restricted. Suited to AI inference workloads.
Export-compliance variant with 17% fewer CUDA cores, 84 GB (vs 96 GB), ~1,398 GB/s bandwidth, and no NVLink versus the standard RTX PRO 6000.
Nvidia RTX PRO 6000
96 GB GDDR7 · released Q1 2025
Nvidia RTX 5090
32 GB GDDR7 · consumer card
Last updated