UseAI glossary
Model routing
Classify each request with a fast, cheap model, then only send to the expensive model what deserves it. On real traffic, the vast majority of requests don't deserve it: the bill is divided without degrading what matters.
In other words
The emergency department reception. Not everyone gets to see the surgeon.
Alyriaz articles that mention it
- Can a model's intelligence be stolen?
Model extraction involves reconstructing a system's capabilities by querying it at scale. We explain the mechanism, the possible defences, and why none of them really works.
- Washington accuses Chinese labs of siphoning intelligence from American models
US agencies accuse Chinese labs of extracting the capabilities of American models at scale. We explain the technique in question and why it is hard to prevent.
- A model that writes a text in one go instead of word by word
Diffusion models applied to text generate an entire response in parallel rather than word by word. We explain the principle, its speed advantages and its limitations.
- ChatGPT Images 2.5: three billion images a week, and a shift towards editing
OpenAI has released ChatGPT Images 2.5 with two models for developers, halved latency, and gains focused on editing. We analyse this shift and what it targets.
- What model size for which use case?
Most tasks don't require the most powerful model. We explain which size suits which use case, and why choosing too large costs more without adding anything.
- What is a model's "effort level"? The setting that changes everything and that nobody looks at
Recent models offer a reasoning-effort setting that profoundly changes their outputs and cost. Here’s what this parameter does, why comparisons depend on it, and how to tune it.