GPU model · price + stock · hourly snapshot
Nvidia GB300 cloud rental prices
1 provider lists the Nvidia GB300, 0 have it in stock. It rents for $13.30 per GPU hour.
LIVE Prices and stock checked 1 minute ago
Providers
Every provider renting the Nvidia GB300
Billing:
Listed, but nobody reports stock at this price type right now.
| Provider | From | Stock | Grab |
|---|---|---|---|
| Verda | $13.30 | Not reported | Grab |
In stock first, then cheapest. Stock is what the provider reported at our last check. One GPU for a month (730h) at $13.30: $9,713.
Price history
Price trend
$13.18 /GPU/hr · Nov 4
One point per day: the lowest listed $/GPU/hr across every provider we track, captured daily. 57 offers in the latest snapshot.
Configurations
All 6 configurations
| Configuration | VRAM | Per GPU/hr | Stock |
|---|---|---|---|
| 4x GB300 SXM6 288GBSpotVerda | 1152 GB | $6.65 | Not reported |
| 1x GB300 SXM6 288GBSpotVerda | 288 GB | $6.65 | Not reported |
| 2x GB300 SXM6 288GBSpotVerda | 576 GB | $6.66 | Not reported |
| 2x GB300 SXM6 288GBVerda | 576 GB | $13.30 | Not reported |
| 4x GB300 SXM6 288GBVerda | 1152 GB | $13.30 | Not reported |
| 1x GB300 SXM6 288GBVerda | 288 GB | $13.31 | Not reported |
Model fit
What fits in 288 GB
The question behind most rentals: will the model load. Sized against one Nvidia GB300.
| Model | 16-bit | 8-bit | 4-bit |
|---|---|---|---|
| DeepSeek-V3 DeepSeek · 671B params · 37B active | 1,610 GB 6× · 118 GB free | 805 GB 3× · 59 GB free | 403 GB 2× · 173 GB free |
| Llama 3.1 405B Meta · 405B params | 972 GB 4× · 180 GB free | 486 GB 2× · 90 GB free | 243 GB 1× · 45 GB free |
| Qwen3 235B-A22B Alibaba · 235B params · 22B active | 564 GB 2× · 12 GB free | 282 GB 1× · 6 GB free | 141 GB 1× · 147 GB free |
| Mixtral 8x22B Mistral · 141B params · 39B active | 338 GB 2× · 238 GB free | 169 GB 1× · 119 GB free | 85 GB 1× · 203 GB free |
| Mistral Large 2 Mistral · 123B params | 295 GB 2× · 281 GB free | 148 GB 1× · 140 GB free | 74 GB 1× · 214 GB free |
| gpt-oss-120b OpenAI · 117B params · 5.1B active | 281 GB 1× · 7 GB free | 140 GB 1× · 148 GB free | 70 GB 1× · 218 GB free |
| Qwen2.5 72B Alibaba · 72.7B params | 174 GB 1× · 114 GB free | 87 GB 1× · 201 GB free | 44 GB 1× · 244 GB free |
| Llama 3.3 70B Meta · 70.6B params | 169 GB 1× · 119 GB free | 85 GB 1× · 203 GB free | 42 GB 1× · 246 GB free |
| Qwen3 32B Alibaba · 32.8B params | 79 GB 1× · 209 GB free | 39 GB 1× · 249 GB free | 20 GB 1× · 268 GB free |
| Gemma 2 27B Google · 27.2B params | 65 GB 1× · 223 GB free | 33 GB 1× · 255 GB free | 16 GB 1× · 272 GB free |
| Mistral Small 3 Mistral · 24B params | 58 GB 1× · 230 GB free | 29 GB 1× · 259 GB free | 14 GB 1× · 274 GB free |
| Llama 3.1 8B Meta · 8B params | 19 GB 1× · 269 GB free | 10 GB 1× · 278 GB free | 5 GB 1× · 283 GB free |
Required memory is weights plus 20% for KV cache and runtime at short context; long contexts and large batches need more. Cells tinted green fit on a single Nvidia GB300.
Specifications
Nvidia GB300 specifications
| Architecture | Grace Blackwell Ultra |
|---|---|
| Memory per GPU | 288 GB HBM3e |
| Memory bandwidth | 8,000 GB/s |
| Release date | Q1 2025 |
| GPU interconnect | NVLink 1,800 GB/s |
| FP4 compute rate | 15,000 TFLOPS (dense) |
| FP8 compute rate | 5,000 TFLOPS (dense), 10,000 with sparsity |
| FP16 / BF16 compute rate | 2,500 TFLOPS (dense), 5,000 with sparsity |
| INT8 compute rate | 165 TOPS (dense), 330 with sparsity |
| Process node | TSMC 4NP |
| Board power | 1,400 W |
| Partitioning | MIG, up to 7 instances |
Alternatives