where thinking models do well. Somewhere in the middle. See, there are three distinct complexity bands: → Low: Non-thinking LLMs actually outscore reasoning models.
→ Medium: The reasoning model's chain-of-thought helps.
→ High: Both vanilla and reasoning models drop to 0 %.
@thatroblennon
-
Three Complexity Bands: Where Reasoning Models Excel
By
–
-
Reasoning Models and the Goldilocks Zone of AI Thinking
By
–
Fast forward to today and we now have reasoning models ("LRMs") that are trained to think. When these first came out, I was skeptical. In my experience, LLM thinking was flawed. Apple's new paper confirms a similar finding. In fact, the identify a "Goldilocks band"…
-
Thought Prompts: Mixed Results in AI Output Quality
By
–
About half of the time, thought prompts worked well — better than without the thinking. And about half the time the thinking process actually made the output WORSE.
-
Thought Prompting Course: Promising Start, Unexpected Challenges
By
–
Last year I started working on a course called "Thought Prompting". I was going to teach people how to prompt LLMs to think before returning the final results. Early experiments showed a ton of promise… but when I started working on the course, things went sideways.
-
Apple’s Paper Confirms Year-Long Observations About Thinking Models
By
–
Apple just dropped an machine learning paper that finally confirms what I've been noticing for over a year. The paper is about "thinking models" like OpenAI's o-series (o3, etc) Here's what happened:
-
Creator Uses Only Custom GPTs, Over 100 Built
By
–
i have over 100 GPTs and use a dozen each week. Zero are other peoples
-
Love the Integration Beyond Agent Platform
By
–
looks sick. I love how it goes outside the agent platform and integrates with everything
-
Marketers Will Continue Using Agent Term for Workflows and GPTs
By
–
exactly. I don't think marketers are going to stop using the term to describe workflow agents and GPTs, so we may as well make room for them in the definition
-
Claude Accidentally Mentioned as Invited Guest During Recording
By
–
little slip up while recording. Claude is indeed invited
-
Introducing curious minds to AI through accessible vibe coding tools
By
–
I feel like this is a good time to expose them to this kind of stuff. They're very curious about AI and what I do, and I love how accessible these vibe coding tools make the whole world of *building*. It's a bit of a new experiment, but so far lots of interest