Controlling Reasoning Effort in LLMs: From OpenAI o1 to GPT-5.6
Reasoning models like OpenAI o1 and DeepSeek-R1 are now standard, using reinforcement learning with verifiable rewards to generate step-by-step traces. I explain how to build models with adjustable reasoning effort settings, similar to the new GPT-5.6 family. We explore training scaling versus inference scaling, debunking myths about special tokens, and show how to balance performance with compute costs.
In the context of AI and LLM research, reasoning model means a model that outputs an intermediate reasoning trace, which is like an intermediate response that works through a question or task step by step.
- simonw
I'm amused by how the whole reasoning model thing feels like a formalization of the old "think step by step" prompting hack, which was discovered against GPT-3 two years after that model was first released.
My favorite trick for controlling the reasoning level is the hack where you look at the output token stream and spot the token for "the model has concluded reasoning"... and then replace that with the tokens for "wait, but" and force it to keep going!
- sva_
I recommend his book, "Build a Reasoning Model (From Scratch)", which is also linked in the article.
https://sebastianraschka.com/books/#build-a-reasoning-model-...
- razorbeamz
LLMs don't actually reason, they just create the illusion of reasoning by repeating things over and over.