All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
On-demand & long-term GPU rental | RTX 4090, 5090, 6000 PRO | H100, H200, GB200 | LLM Inference
NoVibeCoding
Member, Host Rep
Hi LowEndTalk Community!
I am happy to announce CloudRift, a new cloud that allows you to rent GPUs on-demand directly from data center owners.
💎 On-demand GPU rental 💎
We offer on-demand and long-term GPU rental for RTX 4090, RTX 5090, and soon RTX PRO 6000. Check out our offerings at https://console.cloudrift.ai/.
RTX 5090 server specs:
- 100.0 GB RAM per GPU
- 16 CPU cores per GPU (AMD EPYC Milan)
- 1000 GB NVMe storage per GPU
- 1 static IPv4
- Unmetered 10Gbps internet
RTX 4090: $0.40 per hour on-demand | $0.24 per hour for 1-year reservation
RTX 5090: $0.65 per hour on-demand | $0.39 per hour for 1-year reservation
Contact us at [email protected] for longer than 3-month terms or if you're looking for datacenter GPUs like H100, H200, or GB200, which are available for long-term rental. Bare-metal, VM, or Kubernetes.
🙀 Hosted LLM models are 50% off until the end of June 🙀
You can also find hosted popular models like DeepSeek and Llama on the platform, which are 50% off in May and June.
Llama 4 costs $0.1 and $0.3 per 1M input and output tokens, respectively, and DeepSeek V3 costs $0.15 and $0.40.
Enjoy a great deal while our excess GPU capacity lasts.
💰 Sell your data center capacity 💰
If you have a data center operator and are interested in selling GPU compute directly to end-users or LLM Inference, contact us at [email protected]. We are fully transparent with providers on the platform, offering both a marketplace and a white-label solution embeddable on your website.
Comments
Plot twist: the entire platform is built with vibe coding.
It would be convenient... Unfortunately, codegen is not so great with Rust, and I wouldn't trust it with hypervisor implementation.
Congrats and welcome to LET
What does this mean , do you rent datacenters fully stocked racks so they can resell them?
I'm not sure based on what market research did you find LET to be a great target audience for these services. Aren't cheap LET people more likely to just keep using free chetgippity instead of paying per token?
There are two options:
Yes, for chatting and casual use. For app development, ChatGPT's free tier is not enough. Even solo projects run into rate limits very quickly, and the paid version quickly becomes very expensive.