LLM finetuning & inference
Llama, Mistral, Qwen, DeepSeek finetuning with LoRA / QLoRA / full FT on H100. Or self-hosted inference with vLLM / TGI / Ollama for production model serving.
Moldova is the budget GPU tier in our network. Same NVIDIA hardware, but lower electricity cost and minimal regulatory framework let us price GPU plans 10-15% below Romania and 15-25% below Iceland. Use this jurisdiction when cost-per-token matters more than peering or marketing posture.
All plans include CUDA 12 + cuDNN preinstalled, NVMe SSD, DDR5 RAM, full root access, SSH + JupyterLab and unlimited bandwidth.
GPU servers shine on workloads that scale with VRAM and tensor cores — LLM finetuning and inference, diffusion image generation, AI video, and high-throughput model serving.
Strong privacy laws, renewable energy, outside EU.
Courts struck down data retention laws. Great EU connectivity.
Excellent connectivity, tolerant hosting, AMS-IX peering.