The best reasoning models, according to hackers
4 reasoning models recommended in Hacker News comments over the past year, ranked by how many separate people recommend each one.
All reasoning models
- OpenAI o3People recommend o3 for analytical, medical and coding questions, where it gave fewer wrong or nonsensical answers. The common caveat is that o3 Pro is slow and expensive.37recommend
- VibeThinker 3BSmall open model that people recommend for math and logical reasoning, performing well above its size. Common caveats: weak outside its specialty, poor at structured output and tool use.28recommend
- DeepSeek R1Free open-weight reasoning model under an MIT license that people run locally and find capable. Critics point to language mixing, loops, and censorship.25recommend
- OpenAI o1People recommend o1 as an early step-change in AI reasoning on hard problems, though critics say its hype outran its real-world results and its price was high.12recommend