New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Looking for cheap hourly GPU with ready-to-use LLM setup*
Hi guys,
I’m looking for a GPU server for a short batch-processing job.
Requirements:
- Hourly billing
- 24 GB VRAM minimum
- RTX 3090 / 4090 / A5000 or similar
- Preferably 48 GB if the price is good
- Ready-to-use image with CUDA + vLLM or Ollama
- OpenAI-compatible API would be ideal
- Qwen 14B/32B or similar model
- Good disk/network speed for downloading models
- Around 5–15 hours of usage initially
I don’t need a monthly dedicated server or a long-term contract. Just something quick to deploy, run the batch and destroy afterwards.
Any provider offering free trial credit, initial credit or a cheap test would be even better.
I already know RunPod and Vast.ai, but I’m curious if anyone here knows a smaller provider or a good current deal.
Thanks!
Comments
No hourly servers but very cheap, good network: https://www.maghost.com/servers/vps/#vps_gpu-based