Companion videos

Multi-model orchestration — companion videos

The article argues that "use GPT-5/Claude/Gemini for everything" is the wrong default — once you care about cost and latency, you start sending easy work to small models and reserving the big ones for jobs that need them. The first video gives you the mental map for what each tier of model is actually good at; the second shows the routing pattern in code so you can see how it falls out of a normal API call.

Primary pick

19:01

Every AI Model Explained

Tina Huang

A clean tour of the current model landscape grouped by tier — flagships, lite models, mid-tier, specialized — with concrete picks for what each tier is actually good for. This is the "know your options before you route" half of the article, and Huang frames cost-vs-capability the same way the article does without leaning on benchmark hype.

What you should get from this: Compare flagship, lite, mid-tier and specialized models so routing decisions are based on task fit, cost and latency instead of brand preference.

Watch or know first: None beyond the article — it's the know-your-options half before any routing work.

AI Expert note: Model names, pricing and capabilities change quickly. Use this for the decision pattern, then verify current model behavior before adopting it.

Open video page

Also worth watching

08:53

RouteLLM achieves 90% GPT4o Quality AND 80% CHEAPER

Matthew Berman

Walks through the LMSYS RouteLLM paper and code: a small classifier sits in front of a strong/weak model pair and decides which one to call, hitting roughly 95% of the strong model's quality at a fraction of the cost. The view count is under the usual 100k bar, but for the specific "show me real model routing, not just model comparisons" niche this is the cleanest explanation on YouTube and lines up directly with the article's quality/cost tradeoff section.

What you should get from this: Evaluate model-routing tradeoffs between quality, cost and reliability before adding orchestration complexity.

Watch or know first: Comfort reading Python and calling model APIs; the primary pick's tier map helps.

AI Expert note: Model names, pricing and capabilities change quickly. Use this for the decision pattern, then verify current model behavior before adopting it.

Open video page