Best reasoning models

Chain-of-thought / o1-style local thinkers

Models trained to show their work. Ideal for math, code, and multi-step logic puzzles. All run with `<think>` traces enabled.

  1. 1

    DeepSeek

    DeepSeek R1 Distill 8B

    Compact reasoning model. Good reasoning capabilities in a small package.

    8B≥ 5.08 GB
  2. 2

    DeepSeek

    DeepSeek R1 Distill 1.5B

    Compact reasoning model distilled from DeepSeek R1. Strong chain-of-thought in a tiny package.

    1.5B≥ 1.54 GB
  3. 3

    Microsoft

    Phi-4

    Microsoft's 14B parameter model. Punches well above its weight on reasoning.

    14B≥ 8.93 GB
  4. 4

    Alibaba

    Qwen 2.5 32B

    Premium 32B model. Top-tier reasoning. Mac with 32GB+ RAM.

    32B≥ 18.99 GB
  5. 5

    Alibaba

    Qwen 2.5 14B

    Strong 14B model with excellent coding and reasoning. iPad Pro recommended.

    14B≥ 8.87 GB
  6. 6

    Google

    Gemma 3 27B

    Google's flagship open model. Near GPT-4 quality. Needs 20GB+ RAM.

    27B≥ 15.91 GB

Not sure which fits your machine? Auto-detect your hardware →