Entry-level Blackwell professional GPU for AI prototyping.
Weekly median price per GPU per hour
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.
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 4000 has 24 GB of VRAM. In practice, that's enough memory for roughly 28B parameters at 4-bit or 6B 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 4000s 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
|
4
|
–
|
–
|
GLM-5.3-Flash
320B · 18B active
|
178 GB
|
9
|
–
|
–
|
MiniMax-M3
428B · 23B active
|
239 GB
|
12
|
–
|
–
|
DeepSeek V4.1 Flash
552B · 16B active
|
306 GB
|
15
|
–
|
–
|
Kimi K3
2.8T · 104B active
|
1,544 GB
|
72
|
–
|
–
|
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. How we estimate costs.
Nvidia RTX PRO 4000 · Per GPU
| Compute · dense | |
|---|---|
| FP4 | 589 TFLOPS 1,178 with sparsity |
| FP8 | 294.5 TFLOPS 589 with sparsity |
| FP16 / BF16 | 147.3 TFLOPS 294.5 with sparsity |
| INT8 | 294.5 TOPS 589 with sparsity |
| FP32 | 37 TFLOPS |
| Precision support | FP4FP8FP16BF16TF32FP32INT8 |
| Memory | |
|---|---|
| Capacity | 24 GB GDDR7 |
| Bandwidth | 672 GB/s |
| Bus width | 192-bit |
| ECC | Yes |
| Silicon | |
|---|---|
| Architecture | Blackwell |
| Process | TSMC 4N |
| Transistors | 45.6 billion |
| Shader cores | 8,960 CUDA cores |
| Matrix cores | 280 Tensor cores |
| Compute units | 70 SMs |
| Fabric and host | |
|---|---|
| Host interface | PCIe 5.0 x16 |
| Power | |
|---|---|
| Board power | 145 W |
| Cooling | Active |
As of September 13, 2026, we track 33 configs from 3 providers. Prices are per GPU per hour.
| Billing type | Configs | Cheapest |
|---|---|---|
On-demand | 16 | $0.20 (in stock, Vast.ai) |
Reserved | 6 | $0.26 (24 mo) |
Spot | 11 | $0.13 (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 $144 per month on-demand, $194 reserved, $94 spot.
GetDeploying currently tracks RTX PRO 4000 configs from 3 providers. Vast.ai and Runpod have verified in-stock on-demand configs. See the full price comparison above for every provider and config.
One RTX PRO 4000 runs models up to roughly 28B parameters at 4-bit quantization or 6B 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.
24GB GDDR7 with Blackwell FP4 Tensor Cores. Entry-level Blackwell professional GPU for AI prototyping and visualization.
Limited VRAM for larger models. Desktop form factor with minimal cloud availability.
Last updated
