Ultra TechArabic edition

Understand AI tools, and run them yourself.

Full-length guides to AI software and hardware: every fact from its source, every example actually run, every guide checked before it is published.

Start with the first guide

  • 6 guides
  • 70 sources checked
  • Last check

Newest guideSoftware

What Structured Output Is

Write a JSON Schema for the data you need, and send it with a request so the reply follows it.

What changed recently

  1. OllamaOllama 0.40.8 adds Flint, Recursive's model for coding agents, on Macs only for now
  2. OpenAIOpenAI lets you choose which service tiers each API project can use
  3. OllamaOllama 0.40.7 fixes Responses conversations for Codex and lets more draft models speed up replies

Software

All Software guides (4)
  1. What Prompt Caching Is

    Turn on prompt caching by adding two lines to a request, and prove from the response's own numbers that it worked.

  2. What Tool Use Is

    Define one tool, send it with a question, and see the model ask for it instead of guessing — in a file you wrote and ran yourself.

  3. What MCP Is

    Write an MCP server that offers one tool, and send it a message by hand to see exactly what it answers.

Hardware

All Hardware guides (2)
  1. Which Machine Runs a Model Locally

    Read how much memory your machine has, and how much of it belongs to the graphics card, with one file you run yourself.

  2. Mac or PC for Running Models Locally

    List the models you want to run and say how much memory each one needs, with one file you run yourself.

  3. Your First Local Model with Ollama coming

Comparisons

Which RTX 50 card for running models locally

GeForce RTX 5090GeForce RTX 5080GeForce RTX 5070 TiGeForce RTX 5070