This is interesting, but the values could use some refinement:<p><pre><code> gpt-4o: speed 3/5
gpt-4.5-preview: speed 3/5
gpt-3.5-turbo: speed 2/5
</code></pre>
In practice 3.5 turbo is, what, 5-10x <i>faster</i> than 4.5 preview?
The pricing of all these AI models is so opaque.
Can’t they hire someone with common sense who realizes that $x/token is absurd!!!<p>Totally brain-dead. Like pricing a car by the number of o-rings in the engine.