whatmodel.sh / OpenThinker 32B

OpenThinker 32B

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

OpenThinker 32B is an open-weights language model from Open Thoughts, released February 2025. Its published size is 32.76B parameters and a 33k-token context window. Licence: apache-2.0.

Catalogue identity notes

open-thoughts/OpenThinker-32B @840749c (created 2025-02-12, licence apache-2.0). Open Thoughts' first-generation reasoning model fine-tuned from Qwen2.5-32B-Instruct on OpenThoughts-114k (Jan/Feb 2025). Rows from Open Thoughts' Evalchemy tables (own cards and the OpenThinker2 cards), with catalogued comparators (DeepSeek-R1-Distill-Qwen-32B, s1.1-32B).

Model information sources

Release date ↗ · Weights licence ↗

HF safetensors metadata 32,763,876,352 parameters (open-thoughts/OpenThinker-32B @840749c).

54.6±1.3#273 of 719 LLMsfrom 8 benchmark results across 2 sources

$ how it compares

The LLMs ranked either side of OpenThinker 32B, out of 828 with a score, then more from Open Thoughts.

#modelscorewhereRAMreleased
325Bielik 11B v3.0 Instruct SpeakLeash55±2local~8.1GB2025-11
326Olmo 3.1 32B Instruct Ai255±2local~21GB2025-12
327OpenThinker 32B Open Thoughts55±1local~21GB2025-02
328Magistral Small 1.1 (2507) Mistral55±2local~16GB2025-07
329EXAONE Deep 7.8B LG AI Research55±2local~6.5GB2025-03
more from Open Thoughts
331OpenThinker2 32B Open Thoughts54±1local~21GB2025-04
404OpenThinker3 7B Open Thoughts50±1local~6GB2025-05
428OpenThinker2 7B Open Thoughts49±2local~6.4GB2025-04

see all LLMs · find the ones that fit your machine

$ what it's good at

coding▁▁▁▃▆██▇▆▄▃▁ #178 of 458 * · 4.6% of its score · 1 test

Beats Qwen3.5 9B, Devstral 2 (2512), Gemini 2.5 Flash here, though they rank higher overall. Trails QwQ-32B-Preview here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
LiveCodeBench v2 lab68.9% ✎▄··█·▄···▄▄█ beats 5/7medium · avg 59%▇▇ 4.6%at its level · small run
math▁▂▃▅▆▅▄█▄▁▁▁ #180 of 398 · 59% of its score · 3 tests

Beats MiniMax-M2.1, GPT-4.1, DeepSeek V3 0324 here, though they rank higher overall. Trails Phi-4-mini-reasoning 3.8B, OpenThinker3 7B, Olmo 3 7B Think here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
AIME 2024 lab +168.0% ✎▁▁▁▂▁▃▁▅███▃ beats 34/85medium · avg 56%▇▇▇▇▇▇▇▇▇▇▇ 27%sharp · at its level · several runs
MATH-500 lab +190.6% ✎▁▃▂▃▅▇▅▇█▅▂▁ beats 41/84very easy · avg 86%▇▇▇▇▇▇▇ 18%several runs
AIME 2025 lab49.3% ✎▁·▁▁▁▂▅▆█▇▅▁ beats 35/133medium · avg 38%▇▇▇▇▇▇ 15%sharp · at its level
other tests1 not in a task · 35% of its score
testresultvs the others in its rundifficultycountswhy
GPQA Diamond lab +163.5% ✎▁▁▁▁▂▄▅▅███▄ beats 122/310medium · avg 52%▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 35%sharp · at its level · several runs

$ 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 OpenThinker 32B · needs ~19.9 GB · from huggingface.co/lmstudio-community
    chat in the terminal
    llama cli -hf lmstudio-community/OpenThinker-32B-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf lmstudio-community/OpenThinker-32B-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 OpenThinker 32B · needs ~19.9 GB · from huggingface.co/lmstudio-community

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/lmstudio-community/OpenThinker-32B-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run OpenThinker 32B · needs ~20 GB · from ollama.com
    chat in the terminal
    ollama run openthinker:32b

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 OpenThinker 32B · needs ~18.4 GB, Mac only · from huggingface.co/AlejandroOlmedo
    chat in the terminal
    mlx_lm.chat --model AlejandroOlmedo/OpenThinker-32B-4bit-mlx
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model AlejandroOlmedo/OpenThinker-32B-4bit-mlx

Official weights and documentation: huggingface.co/open-thoughts ↗

$ how the score was worked out · open-thoughts/OpenThinker2-32… 60%, open-thoughts/OpenThinker-32B… 40%

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
open-thoughts/OpenThinker2-32B model card · lab-reported4▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 60%Apache-2.0
open-thoughts/OpenThinker-32B model card · lab-reported4▇▇▇▇▇▇▇▇▇▇▇▇ 40%Apache-2.0
prior · pull toward the middle▇ 1%
weighted average= 54.6

$ 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