Grace CPU + Blackwell Ultra GPU superchip built for rack-scale AI reasoning and inference.
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 GB300 has 288 GB of VRAM. In practice, that's enough memory for roughly 460B parameters at 4-bit or 125B 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) | GB300s 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
|
1
|
–
|
–
|
MiniMax-M3
428B · 23B active
|
239 GB
|
1
|
–
|
–
|
DeepSeek V4.1 Flash
552B · 16B active
|
306 GB
|
4
|
–
|
–
|
Kimi K3
2.8T · 104B active
|
1,544 GB
|
6
|
–
|
–
|
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 GB300 · Per GPU
| Compute · dense | |
|---|---|
| FP4 | 15,000 TFLOPS |
| FP8 | 5,000 TFLOPS 10,000 with sparsity |
| FP16 / BF16 | 2,500 TFLOPS 5,000 with sparsity |
| INT8 | 165 TOPS 330 with sparsity |
| FP32 | 80 TFLOPS |
| FP64 | 1.3 TFLOPS |
| Precision support | FP4FP6FP8FP16BF16TF32FP32FP64INT8 |
| Memory | |
|---|---|
| Capacity | 288 GB HBM3e |
| Bandwidth | 8,000 GB/s |
| Bus width | 8,192-bit |
| ECC | Yes |
| Silicon | |
|---|---|
| Architecture | Blackwell Ultra |
| Process | TSMC 4NP |
| Transistors | 208 billion |
| Shader cores | 20,480 CUDA cores |
| Matrix cores | 640 Tensor cores |
| Compute units | 160 SMs |
| Fabric and host | |
|---|---|
| GPU interconnect | NVLink 1,800 GB/s |
| Host interface | PCIe 6.0 x16 |
| Power | |
|---|---|
| Board power | 1,400 W |
| Cooling | Liquid |
| Platform | |
|---|---|
| Partitioning | MIG, up to 7 instances |
Source: official Nvidia GB300 datasheet.
As of October 22, 2026, we track 28 configs from 8 providers. Prices are per GPU per hour.
| Billing type | Configs | Median /GPU/hr | Cheapest |
|---|---|---|---|
On-demand | 4 | $18.00 (not verified in stock, Oracle Cloud) |
|
Reserved | 16 | On request | |
Spot | 3 | Sold out |
No median for on-demand: we only show this when at least 3 providers list the GPU on that billing type. 5 providers also quote the GB300 on a custom contract, priced per deal.
The lowest listed on-demand price for the GB300 is $18.00 per GPU per hour from Oracle Cloud, though we haven't verified current stock.
GetDeploying currently tracks GB300 configs from 8 providers. See the full price comparison above for every provider and config.
One GB300 runs models up to roughly 460B parameters at 4-bit quantization or 125B 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.
Pairs Grace CPUs with Blackwell Ultra GPUs carrying 288GB HBM3e each, connected in a 72-GPU NVLink domain. Suited for frontier-scale training and long-context reasoning inference where per-GPU memory and interconnect bandwidth are the bottleneck.
Last updated
