Every host for Nemotron 3 Ultra
| Host | Blended | Input | Output | Cached | Precision | Uptime |
|---|---|---|---|---|---|---|
| DeepInfra | $0.74 | $0.40 | $1.76 | $0.08 | fp4 | 100.0% |
| Perplexity | $0.813 | $0.25 | $2.50 | $0.25 | — | — |
| Baseten | $1.05 | $0.60 | $2.40 | $0.12 | fp4 | 99.9% |
| Venice | $1.25 | $0.625 | $3.13 | $0.188 | fp8 | 99.9% |
| Together | $1.35 | $0.60 | $3.60 | $0.20 | — | — |
| Nebius | $1.50 | $1.00 | $3.00 | — | — | — |
Cheapest price over time
- Input
- Output
What your workload costs
- DeepInfra$296.00
- Perplexity$325.00
- Baseten$420.00
- Venice$500.00
- Together$540.00
- Nebius$600.00
Rent GPUs or pay per token?
- Renting the GPUs
- $16,878/mo
- Break-even volume
- 22.8B tokens/mo
- Needed around the clock
- 8.7K tok/s
8 × H100 cost $16,878 a month. The cheapest API (DeepInfra) charges $0.74 per 1M blended tokens, so the GPUs only pay off above 22.8B tokens a month: the box must serve 8.7K tokens a second, every second.
GPU price: cheapest on-demand offer per GPU-hour × 730 hours. No throughput is assumed: it depends on the model size, precision, batch size and serving software, so measure yours. Power, storage, egress and engineering time are not included.
More from NVIDIA
Quality data: LMArena (CC BY 4.0), Epoch AI (CC BY 4.0). Prices: OpenRouter, LiteLLM (MIT), provider pages. List prices in US dollars before discounts and taxes; other currencies converted at the European Central Bank reference rate. The figures on these pages are free to reuse under CC BY 4.0 with a link to this page.