whatmodel.sh / K2-Think

K2-Think

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

K2-Think is an open-weights language model from MBZUAI IFM (LLM360), released September 2025. Its published size is 32.76B parameters. Licence: apache-2.0.

Catalogue identity notes

Maker repo IFM/K2-Think @9e218ac (requested as LLM360/K2-Think, which the HF API redirects to IFM/K2-Think): licence tag apache-2.0. 32,763,876,352 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2025-09: month the maker's Hugging Face repo was created (createdAt 2025-09-08).

55.7±4.4#299 of 949 LLMsfrom 1 benchmark result across 1 source* thinly tested, treat the rank as provisional

$ how it compares

The LLMs ranked either side of K2-Think, out of 1132 with a score.

#modelscorewhereRAMreleased
395Qwen-Max (2024-09-19) Alibaba56±3comparison only2024-09
396Hunyuan-Large-Vision Tencent56±3comparison only2025-05
397K2-Think MBZUAI IFM (LLM360)56±4local~21GB2025-09
398FuseChat-Llama-3.2-3B-Instruct FuseAI56±2local~3.4GB2024-12
399Yehia 7B Reasoning (preview) JasperV1356±4local~16GB2025-04

see all LLMs · find the ones that fit your machine

$ what it's good at

other tests1 not in a task · 91% of its score
testresultvs the others in its rundifficultycountswhy
BRIDGE clinical text, zero-shot average (87 tasks)37.8% ✎▁▁▁▁▁▂▂▃▄▆█▃ beats 59/97hard · avg 32%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 91%blunt · at its level

$ where to run it

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run K2-Think · needs ~19.9 GB · from huggingface.co/mradermacher
    chat in the terminal
    llama cli -hf mradermacher/K2-Think-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf mradermacher/K2-Think-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run K2-Think · needs ~19.9 GB · from huggingface.co/mradermacher

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/mradermacher/K2-Think-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run K2-Think · needs ~19.9 GB · from huggingface.co/mradermacher
    chat in the terminal
    ollama run hf.co/mradermacher/K2-Think-GGUF:Q4_K_M

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run K2-Think · needs ~18.4 GB, Mac only · from huggingface.co/qurk41
    chat in the terminal
    mlx_lm.chat --model qurk41/K2-Think-mlx-4Bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model qurk41/K2-Think-mlx-4Bit

Other ways to run it

  1. Run it as published · needs ~65.53 GB · from huggingface.co/IFM (official)
    uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='IFM/K2-Think', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"

Official weights and documentation: huggingface.co/IFM ↗

$ how the score was worked out · BRIDGE Medical Leaderboard (H… 91%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
BRIDGE Medical Leaderboard (Harvard Medical School / Brigham and Women's YLab)1▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 91%MIT
prior · pull toward the middle▇▇▇ 9%
weighted average= 55.7

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json