Yes, that (and some other similar issues) is why the era is coming to a close — it's no longer sufficient to rely on it in practice, since costs are getting too high vs improvements.
@jeremyphoward
-
Grok 3 Comparison: AI Model Performance Assessment
By
–
"Better than Grok 3" — damning with faint praise? 😉
-
Custom Qwen Inference Servers Hardware Cost Analysis
By
–
Well we have our own inference servers and custom config for qwen, so it's really the cost of that hardware I guess.
-
In-house Everything Strategy: Building Complete Internal Infrastructure
By
–
We have our own inhouse-everything 🙂
-
Grok-4 vs Grok-3: Quality Comparison Analysis
By
–
There is indeed a big difference between grok-3 and grok-4. But… it isn't because grok-4 is really good.
-
Deep Research provides significantly more information than standard prompt
By
–
Answer is definitely "no". Here's the same prompt in Deep Research – much much more information here! https://
chatgpt.com/share/68980afd
-2d50-800c-b23b-af9b7043f692
… -
Automatic routing versus deep thinking in AI systems
By
–
No, that's if you are "automatically routed", not if you ask to "think deep".
-
Deep Research Model Outdated Compared to GPT 5
By
–
Looks like Deep Research is still using an older model? GPT 5 knows about FastHTML (which is a year old), but if I select "Deep research" then it doesn't.
-
ChatGPT Deep Research Feature Comparison Analysis
By
–
Output reminds me a bit of Deep Research… https://
chatgpt.com/share/689803ff
-6330-800c-8947-c8d0fb26cf57
…