Grace CPU + Blackwell GPU superchip with unified memory architecture.
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 GB200 has 186 GB of VRAM. In practice, that's enough memory for roughly 293B parameters at 4-bit or 79B 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) | GB200s needed | Cost /hr | Cost /mo |
|---|---|---|---|---|
Qwen3.8-27B
27B
|
18 GB
|
1
|
$16.00
|
$11,520
|
Gemma 4 31B
30.7B
|
22 GB
|
1
|
$16.00
|
$11,520
|
GPT-OSS-120B
117B · 5.1B active
|
67 GB
|
1
|
$16.00
|
$11,520
|
GLM-5.3-Flash
320B · 18B active
|
178 GB
|
4
|
$64.00
|
$46,080
|
MiniMax-M3
428B · 23B active
|
239 GB
|
4
|
$64.00
|
$46,080
|
DeepSeek V4.1 Flash
552B · 16B active
|
306 GB
|
4
|
$64.00
|
$46,080
|
Kimi K3
2.8T · 104B active
|
1,544 GB
|
10
|
–
|
–
|
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 GB200 · Per GPU
| Compute · dense | |
|---|---|
| FP4 | 10,000 TFLOPS 20,000 with sparsity |
| FP8 | 5,000 TFLOPS 10,000 with sparsity |
| FP16 / BF16 | 2,500 TFLOPS 5,000 with sparsity |
| INT8 | 5,000 TOPS 10,000 with sparsity |
| FP32 | 80 TFLOPS |
| FP64 | 40 TFLOPS |
| Precision support | FP4FP6FP8FP16BF16TF32FP32FP64INT8 |
| Memory | |
|---|---|
| Capacity | 186 GB HBM3e |
| Bandwidth | 8,000 GB/s |
| ECC | Yes |
| Silicon | |
|---|---|
| Architecture | Blackwell |
| Process | TSMC 4NP |
| Transistors | 208 billion |
| Fabric and host | |
|---|---|
| GPU interconnect | NVLink 1,800 GB/s |
| Host interface | PCIe 5.0 x16 |
| Power | |
|---|---|
| Board power | 1,200 W |
| Cooling | Liquid |
| Platform | |
|---|---|
| Partitioning | MIG, up to 7 instances |
Source: official Nvidia GB200 datasheet.
As of October 1, 2026, the median on-demand price is $16.00 per GPU per hour across 3 providers.
| Billing type | Configs | Median /GPU/hr | Cheapest |
|---|---|---|---|
On-demand | 3 | $16.00 | $10.50 |
Reserved | 3 | $10.58 |
|
Spot | 1 | $27.04 |
No median for reserved and spot: we only show this when at least 3 providers list the GPU on that billing type. 6 providers also quote the GB200 on a custom contract, priced per deal.
The lowest listed on-demand price for the GB200 is $10.50 per GPU per hour from CoreWeave, though we haven't verified current stock.
At 720 hours per month, one GB200 costs an estimated $11,520 at the median on-demand price.
GetDeploying currently tracks GB200 configs from 10 providers. See the full price comparison above for every provider and config.
One GB200 runs models up to roughly 293B parameters at 4-bit quantization or 79B 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.
Combined Grace CPU + Blackwell GPU with 186GB HBM3e and 1800 GB/s NVLink. Unified memory architecture eliminates CPU-GPU bottlenecks for large-scale AI training.
Specialized superchip form factor with limited cloud availability. The standalone B200 is more widely offered and sufficient for most Blackwell workloads.
Nvidia H100
80 GB HBM3 · 58% less bandwidth than the GB200
Nvidia B200
180 GB HBM3e · released Q1 2024
Nvidia GB300
288 GB HBM3e · released Q1 2025
Nvidia GH200
96 GB HBM3 · 50% less bandwidth than the GB200
Last updated
