Is a reasoning model just a bigger model?

No. Model size and reasoning capability are separate axes. A reasoning model can be built on a smaller base model and still outperform a larger non-reasoning model on tasks that need multi-step logic.

Can I see the reasoning trace?

It depends on the provider. Some return a summary of the trace. Others keep it hidden and hand back only the final answer, for product and safety reasons.

MODEL TYPE

What Is a Reasoning Model?

A reasoning model is trained to work through a long run of intermediate steps before it gives a final answer, instead of replying straight away. That buys better accuracy on tasks with multi-step logic, and it costs more compute and latency per query.

Two-lane diagram: on the left a model answers a word problem instantly and gets 42; on the right the same model works through each step and gets 56. A dial below controls how hard the model thinks. — One-pager: the same word problem answered instantly (wrong) and worked through step by step (right).

Walkthrough: with and without thinking mode on the same problem, plus the one dial that controls it.

What the video explains

The same problem, with and without thinking mode

Say you ask a model this. A monkey eats three bananas a day. Its friend eats two more bananas per day than that. Together, how many bananas do they eat in a week? Without thinking mode, the model answers right away, and it gets this wrong. It says 42. It skipped a step.

With thinking mode on, the model gets a budget to work through the same problem first. It works out each monkey's daily total before answering. This time it lands on the right answer, 56.

One dial, and what skipping it costs

One dial controls how hard the model thinks. Set it low, and it answers faster and cheaper. Set it to extra high, and it's slower and costs more, but it gets more room to work through the problem.

Without thinking mode, a model struggles on multi-step math and logic, on long chains of instructions, and on catching its own mistakes. Same question, same model. The only difference is whether it got room to think first.

What's different

A standard model predicts the next token straight toward its answer. A reasoning model is trained, often with reinforcement learning on verifiable tasks like math and code, to first lay out a chain of steps and then answer from that chain.

This is not the same as telling a standard model to 'think step by step.' That prompt, chain-of-thought, nudges an existing model to show some working in its output. A reasoning model has the behavior trained in, with far longer and more thorough traces than a prompt alone produces.

What it costs

Reasoning tokens cost time and money. A reasoning model can burn far more tokens on the trace, hidden or visible, than on the answer itself, and that lands straight in your latency and your API bill. Providers usually price reasoning tokens the same as output tokens.

The payoff shows up on tasks with a checkable right answer and real multi-step structure: math, code, logic puzzles, planning. On casual conversation or a simple lookup, the extra deliberation adds cost and no accuracy.

When to reach for one

Reach for a reasoning model when a task has several dependent steps and one wrong step breaks the whole answer: multi-step math, debugging, planning a sequence of tool calls. Use a standard model for retrieval, formatting, summarization, and conversation, where the extra thinking changes the bill and not the outcome.

FAQ

Is a reasoning model just a bigger model?: No. Model size and reasoning capability are separate axes. A reasoning model can be built on a smaller base model and still outperform a larger non-reasoning model on tasks that need multi-step logic.
Can I see the reasoning trace?: It depends on the provider. Some return a summary of the trace. Others keep it hidden and hand back only the final answer, for product and safety reasons.

Sources

OpenAI, 'Learning to Reason with LLMs' (o1 system card)

Last updated 2026-09-06