Blog
GPU guides and engineering notes
How much memory a model really needs, which GPU to pick, what a month of compute costs — written by the engineers who run our servers.
Guides · Featured
How much VRAM do I need for an LLM?
How much VRAM do I need for an LLM? Weights by precision, the KV cache formula with worked Llama 3.1 examples, fine-tuning memory and which GPU fits.
- Hardware AMD MI355X for AI: specs, ROCm and the workloads it suits AMD MI355X for AI, explained: 288 GB of HBM3E at 8 TB/s, FP8 and MXFP4, the real state of ROCm, and the workloads that benefit most. Read before you buy.
- Guides Best GPU for LLM inference in 2026: a sizing guide by model size The best GPU for LLM inference depends on model size, precision, KV cache and bandwidth. Use our 2026 sizing table, from RTX 5090 to 8× B300, to choose.
- Guides Dedicated vs shared GPU: what changes when you share Dedicated vs shared GPU: how vGPU, MIG, time-slicing and shared hosts affect memory, consistency, security and debugging, and when sharing is fine.
- Guides DeepSeek R1 hardware requirements: GPUs and memory for R1 and V3 DeepSeek R1 hardware requirements: about 671 GB of FP8 weights plus KV cache. See which 8-GPU servers fit R1 and V3, and the GPUs for distilled models.
- Engineering FP8 vs FP4 vs BF16: number formats for AI, explained FP8 vs FP4 vs BF16 explained: bit layouts, NVFP4 and MXFP4 block scaling, which GPUs support each format, the memory saved and the accuracy you trade.
- Market GPU shortage 2026: what buyers face and how to secure GPUs GPU shortage 2026: on-demand stock sells out, providers push 1-year contracts and prices are up. Why it happens, and how teams still secure GPUs.
- Market H100 rental cost in 2026: what you actually pay H100 rental cost in 2026: SXM vs PCIe, monthly vs hourly billing, egress and storage, plus a break-even formula to compare any quote before you sign.
- Guides How to choose a GPU server for AI How to choose a GPU server for AI: a framework for memory, bandwidth, precision, interconnect, CPU and storage, software and billing, plus a checklist.
- Engineering How to fine-tune a 70B model: GPU requirements and workflow How to fine-tune a 70B model: the memory math for full fine-tuning, LoRA and QLoRA, the GPUs each one needs, and a practical workflow from data to serving.
- Engineering How to self-host LLM inference with vLLM Self-host LLM inference with vLLM: size the GPU, run an OpenAI-compatible server, tune FP8, prefix caching and parallelism, and secure it for production.
- Hardware NVLink vs PCIe: bandwidth, topology and when it matters NVLink vs PCIe compared: per-GPU bandwidth, NVSwitch and NVL72, which multi-GPU workloads need NVLink and which run fine on PCIe, with worked numbers.
Talk to an engineer
Get your GPU server this month.
Tell us what you need. A GPU engineer replies with availability, a confirmed delivery date and a payment link.
- Delivery date confirmed before you pay
- Cancel any month
- Card or crypto, same price