rtx 4090

3 articles · newest first

A high-end multi-GPU AI workstation build

the local AI workstation that runs 70B models: the ~$3,500 build, and the smarter way to spend it

To run a 70-billion-parameter model at home you need ~48GB of VRAM. Here's the build that gets you there — and why two cheaper cards beat one expensive one.

Ravi Malhotra · Jul 30 · 9 min

A cloud GPU workstation for fine-tuning AI models

the best cloud GPU for fine-tuning an LLM costs about a dollar — here's the one to rent and why

Fine-tuning sounds expensive until you realize you can rent the perfect GPU for it by the hour. For small models a 4090 does it for pennies; for bigger jobs an A100 is the workhorse. Here's the cheapest path.

Lena Fischer · Jul 28 · 9 min

Two NVIDIA graphics cards, representing a dual-GPU build for Llama 70B

the honest answer to 'what GPU runs Llama 70B': not the one you were about to buy

Llama 70B needs about 40GB of VRAM to run well — and almost no single consumer card has it. Here's what actually works, visualized, with the measured speeds and real costs for each path.

Ravi Malhotra · Jul 23 · 10 min