[whatmodel.sh](../../README.md) / Ministral 3 14B Reasoning 2512 # Ministral 3 14B Reasoning 2512 Mistral · released 2025-12 · 13.9B params · apache-2.0 **score 56.6** ±0.9 · #251 of 732 · from 20 benchmark results across 6 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at _{● marks Ministral 3 14B Reasoning 2512 among every model on that task or test, worst to best.}

task            vs the others          rank  counts  tests
agent tasks     ▁▁▃●█▄▅▅▄▄▂▁   #250 of 313*    3.0%      2
coding          ▁▁▁▃▅▇●▇▅▄▃▁    #235 of 469     10%      2
chat & writing  ▁▁▁▃▆●▆█▆▇▇▃   #309 of 402*    0.8%      1
math            █·····█····●        #1 of 3     26%      2

agent tasks · #250 of 313 · 3.0% of its score

Beats [o3-mini](o3-mini.md), [Llama 4 Maverick (402B-A17B)](llama-4-maverick.md), [DeepSeek R1 0528 Qwen3 8B](deepseek-r1-0528-qwen3-8b.md) here, though they rank higher overall. Trails [Granite 4.0 1B](granite-4-0-1b.md), [SmolLM3 3B](smollm3-3b.md), [Qwen3-VL 2B Instruct](qwen3-vl-2b-instruct.md) here, though they rank lower overall.

test                             result  vs the others in its run   difficulty        counts  why
BFCL V4 (December 2025 archive)   35.9%  ▁▁▁▂▆▆●▆█▂▄▃ beats 43/114  hard · avg 31%      2.1%  at its level
BFCL v3                           52.2%  ▁▁▁▁▁▃●█▆▅▄▁ beats 50/171  medium · avg 57%    0.9%  blunt · at its level

coding · #235 of 469 · 10% of its score

Beats [Qwen3.5 4B](qwen3-5-4b.md), [Qwen3-VL 32B Instruct](qwen3-vl-32b-instruct.md), [Solar Open 100B (102B-A12B)](solar-open-100b.md) here, though they rank higher overall. Trails [QwQ-32B-Preview](qwq-32b-preview.md), [Ministral 3 3B Reasoning 2512](ministral-3-3b-reasoning.md), [GPT-4 (0613)](gpt-4-0613.md) here, though they rank lower overall.

test                                result  vs the others in its run  difficulty        counts  why
LiveCodeBench (version unspecifie…   64.6%  ▁···▁▁▁█▇●▄▁ beats 29/38  medium · avg 49%    6.6%  at its level · small run · several runs
LiveCodeBench v6 (window unspecif…   42.7%  ▂▃▄●▅██▆▄▇▄▆ beats 12/67  medium · avg 42%    3.0%  at its level

chat & writing · #309 of 402 · 0.8% of its score

Trails [LFM2 350M](lfm2-350m.md), [Gemma 3 1B](gemma-3-1b.md), [Qwen3 0.6B](qwen3-0-6b.md) here, though they rank lower overall.

test                              result  vs the others in its run   difficulty      counts  why
IFEval (aggregation unspecified)   59.7%  ▁▂▄●▂▄▆██▅▂▁ beats 21/145  easy · avg 75%    0.8%  noisy

math · #1 of 3 · 26% of its score

test               result  vs the others in its run   difficulty      counts  why
AIME 2025 (maker)   85.0%  ▁·▁▁▁▂▅▆█●▅▁ beats 99/133  easy · avg 67%     14%  sharp · small run · several runs
AIME 2024 (maker)   89.8%  ▁▁▁▂▁▃▁▅██●▃ beats 72/85   easy · avg 75%     12%  sharp · small run · several runs

other tests (8)

test                                result  vs the others in its run    difficulty        counts  why
GPQA Diamond                         46.0%  ▁▁▁▁▂▄▅▅●██▄ beats 161/322  medium · avg 53%     24%  sharp · at its level · small run · seve…
Medmarks-V mean win rate (19 veri…  0.4807  ▁▄▃▃▇●█▃▅▁▁▁ beats 14/30    avg 0.418            16%  sharp
Medmarks-V weighted mean win rate…   48.1%  ▂▃▅▂▆▃●▅█▅▃▆ beats 13/34    medium · avg 41%    7.0%  blunt · at its level
EuroEval Life in the UK (English …   93.2%  ▂·▂▂·▂·▃▃·●▃ beats 10/14    easy · avg 80%      4.6%  typical
EuroEval HellaSwag (English), val…   86.9%  ▂··▂▂·▂▆▆█·● beats 14/15    medium · avg 63%    3.2%  at its level
MMLU-Redux                           84.4%  ▁▁▃▁▂▃●▅▇▅█▄ beats 17/63    easy · avg 76%      3.1%  typical
SurgiQ surgical MCQ (zero-shot, l…   63.5%  ▂···▂▂·▃▂▃█● beats 13/17    medium · avg 53%    2.1%  at its level
GSM-Plus                             86.5%  ▁··▁▁▁·▃▁●█▁ beats 8/16     easy · avg 84%      1.4%  blunt

## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's 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 is added up by who ran the tests. [Method](../method.md)

source                                    results  share of the score
mistralai/Ministral-3-3B-Reasoning-2512…        4   26% ▇▇▇
mistralai/Ministral-3-8B-Reasoning-2512…        4   26% ▇▇▇
Medmarks: A Comprehensive Open-Source L…        2   23% ▇▇▇
JetBrains/Mellum2-12B-A2.5B-Thinking mo…        7   15% ▇▇
EuroEval leaderboards                           2    8% ▇
SurgiQ: A Large-Scale Multi-Domain Benc…        1    2% ▇
prior: pull toward the middle                        0% ▇
weighted average                                   = 56.6

_{Licences: mistralai/Ministral-3-3B-Reasoning-2512 model card ([Apache-2.0](../../licenses/cards-gap-NOTICE.txt)) · mistralai/Ministral-3-8B-Reasoning-2512 model card ([Apache-2.0](../../licenses/cards-gap-NOTICE.txt)) · Medmarks: A Comprehensive Open-Source LLM Benchmark Suite for Medical Tasks ([CC BY 4.0](../../licenses/paper-2605.01417v1-CC-BY-4.0.txt)) · JetBrains/Mellum2-12B-A2.5B-Thinking model card ([Apache-2.0](https://huggingface.co/JetBrains/Mellum2-12B-A2.5B-Thinking/blob/a7311550557e93cc706ab5dd3d879c1a11703ab4/README.md)) · EuroEval leaderboards ([MIT](https://github.com/EuroEval/leaderboards/blob/1a77f1d2f8ba1ba05ab73287fb090d76bf41b2e0/LICENSE)) · SurgiQ: A Large-Scale Multi-Domain Benchmark for Evaluating Surgical Understanding in Large Language Models ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/))}

Not counted (4)

source                  test                            score  why
mistralai/Ministral-3…  AIME 2025                       85.0%  only model in this source's run
mistralai/Ministral-3…  AIME 2024                       89.8%  only model in this source's run
mistralai/Ministral-3…  GPQA Diamond                    71.2%  only model in this source's run
mistralai/Ministral-3…  LiveCodeBench (version unspec…  64.6%  only model in this source's run

## How to run it ### llama.cpp (recommended) 1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md) ```sh # macOS / Linux curl -LsSf https://llama.app/install.sh | sh # Windows irm https://llama.app/install.ps1 | iex ``` 2. Run Ministral 3 14B Reasoning 2512 (Q4_K_M, needs 8.24 GB, from [huggingface.co/mistralai](https://huggingface.co/mistralai/Ministral-3-14B-Reasoning-2512-GGUF/blob/fe3b038f30334729263d860d5dadbaa34e0f2a18/Ministral-3-14B-Reasoning-2512-Q4_K_M.gguf)) ```sh llama cli -hf mistralai/Ministral-3-14B-Reasoning-2512-GGUF:Q4_K_M ``` 3. Or start a local server and open the address it prints (http://localhost:9931) ```sh llama serve -hf mistralai/Ministral-3-14B-Reasoning-2512-GGUF:Q4_K_M ``` ### LM Studio 1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download) 2. Run Ministral 3 14B Reasoning 2512 (Q4_K_M, needs 8.24 GB, from [huggingface.co/mistralai](https://huggingface.co/mistralai/Ministral-3-14B-Reasoning-2512-GGUF/blob/fe3b038f30334729263d860d5dadbaa34e0f2a18/Ministral-3-14B-Reasoning-2512-Q4_K_M.gguf)) ```sh lms get https://huggingface.co/mistralai/Ministral-3-14B-Reasoning-2512-GGUF@Q4_K_M -y ``` 3. Then run lms ls to see its name, and lms chat followed by that name Other versions: 4bit (~8.4 GB, Mac only) `lms get https://huggingface.co/hunterbown/Ministral-3-14B-Reasoning-2512-MLX-4bit -y` ### Ollama 1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download) ```sh # Linux curl -fsSL https://ollama.com/install.sh | sh ``` 2. Run Ministral 3 14B Reasoning 2512 (Q4_K_M, needs 8.24 GB, from [huggingface.co/mistralai](https://huggingface.co/mistralai/Ministral-3-14B-Reasoning-2512-GGUF/blob/fe3b038f30334729263d860d5dadbaa34e0f2a18/Ministral-3-14B-Reasoning-2512-Q4_K_M.gguf)) ```sh ollama run hf.co/mistralai/Ministral-3-14B-Reasoning-2512-GGUF:Q4_K_M ``` ### MLX (Mac) 1. Install MLX (Mac). Apple silicon Macs only, text only. ```sh # with uv uv tool install mlx-lm # or pip install mlx-lm ``` 2. Run Ministral 3 14B Reasoning 2512 (4bit, needs ~8.4 GB, Mac only, from [huggingface.co/hunterbown](https://huggingface.co/hunterbown/Ministral-3-14B-Reasoning-2512-MLX-4bit/tree/885cdcb8abdf453ac743e6dfb68263c39b5bed35)) ```sh mlx_lm.chat --model hunterbown/Ministral-3-14B-Reasoning-2512-MLX-4bit ``` 3. Or start an OpenAI-compatible server at http://localhost:8080/v1 ```sh mlx_lm.server --model hunterbown/Ministral-3-14B-Reasoning-2512-MLX-4bit ``` Model repository and documentation: [huggingface.co/mistralai](https://huggingface.co/mistralai/Ministral-3-14B-Reasoning-2512) [interactive version](https://whatmodel.sh/model/ministral-3-14b-reasoning) · [JSON](https://whatmodel.sh/api?format=json&all=1)