Open compute, explained
Run the model. Understand the machine.
Current, practical answers for choosing open models, sizing GPUs and operating inference without turning every decision into a research project.
Concepts
4 min read
How to list an idle GPU for rental on Compute-AR
What GPU owners need to provide, how approval works, and what to agree before their hardware is rented.
Models
7 min read
Which open model should you run in 2026?
A practical shortlist for agents, coding, multimodal work, and smaller private deployments.
Hardware
6 min read
How much GPU memory does an LLM need?
Estimate model weights, context cache, runtime overhead, and concurrency before choosing a GPU.
Deployment
5 min read
How to run an open model with Ollama
The quickest path from a model name to a local API on macOS, Windows, or Linux.
Deployment
7 min read
How to serve an open model with vLLM
Launch an OpenAI-compatible inference endpoint and test it before exposing it to an application.
Concepts
6 min read
Ollama vs vLLM vs llama.cpp
Choose an inference stack by environment, concurrency, hardware, and operational burden.
Concepts
8 min read
LLM quantization: 4-bit, 8-bit, GGUF, AWQ and GPTQ
What lower precision changes, when it saves money, and what to test before production.
Concepts
6 min read
RAG vs fine-tuning vs LoRA
Decide whether the model needs new facts, new behaviour, or both.