Blog
Technical writing on GPU rental, pricing, and AI infrastructure. Posts are written by the CoreHu team.
Renting vs. buying a GPU: which one, when?
We compare capital cost, utilization, and hardware depreciation in one equation, and work out the break-even point and the hidden costs.
Interruptible GPU prices: risk and reward
Why is interruptible GPU rental so much cheaper than the standard rate? We examine the interruption risk, the expected-value equation, and how to prepare your workload.
Choosing a GPU for LLM fine-tuning (a VRAM guide)
From 7B to 70B: we calculate the VRAM needed for full fine-tuning and LoRA, and explain how batch size, sequence length, and quantization affect the budget.
Why is crypto mining banned while compute jobs are allowed?
Two workloads on the same GPU: one is banned on almost every platform, the other is the core, contract-governed business. The difference comes down to efficiency, transparency, and abuse risk.
The posts on this page are sample content; they are marked as unpublished drafts until the CMS is connected.