whatmodel.sh / Ministral 3 3B Reasoning 2512

Ministral 3 3B Reasoning 2512

+ contribute

Ministral 3 3B Reasoning 2512 is an open-weights language model from Mistral, released December 2025. Its published size is 3.8B parameters and a 262k-token context window. Licence: apache-2.0.

Model information sources

· Parameter count ↗

Sum of the two components the card states (3.4B LM + 0.4B vision encoder = 3.8B). The repo's safetensors header gives 3,849,090,048 BF16 parameters (HF API); the sibling non-reasoning catalogue entry uses 3.85. Dense model: active_b = params_b.

50.9±1.0#326 of 719 LLMsfrom 12 benchmark results across 4 sources

$ how it compares

The LLMs ranked either side of Ministral 3 3B Reasoning 2512, out of 828 with a score, then more from Mistral.

#	model	score	where	RAM	released
392	Command A Translate (08-2025) Cohere	51±5	api price unknown		2025-08
393	Llama 3.1 Nemotron 51B Instruct NVIDIA	51±2	local	~34GB	2024-09
394	Ministral 3 3B Reasoning 2512 Mistral	51±1	local	~3.5GB	2025-12
395	Qwen3 4B Alibaba	51±1	local	~3.7GB	2025-04
396	Yi-Large 01.AI	51±1	comparison only		2024-05
	more from Mistral
79	Mistral Large 4 (public preview) Mistral	71±2	api $2.06/M		2026-10
174	Mistral Medium 3.5 (128B) Mistral	65±1	api $3.00/M +local	~42GB	2026-04
180	Mistral Medium 3.1 (2508) Mistral	64±2	comparison only		2025-08

see all LLMs · find the ones that fit your machine

$ what it's good at

coding▁▁▁▃▆██▇▆▄▃▁ #195 of 458 * · 10% of its score · 1 test

Beats Gemini 2.5 Flash, Qwen3.5 4B, Qwen3 30B-A3B Instruct 2507 here, though they rank higher overall. Trails QwQ-32B-Preview here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
LiveCodeBench (version unspecified) lab +1	54.8% ✎	▁···▁▁▃▅▅█▄▅ beats 20/37	medium · avg 47%	▇▇▇▇ 10%	at its level · small run · several runs

math█·····█····█ #3 of 3 · 43% of its score · 2 tests

test	result	vs the others in its run	difficulty	counts	why
AIME 2024 lab +1	77.5% ✎	▁▁▁▂▁▃▁▅███▃ beats 50/85	easy · avg 73%	▇▇▇▇▇▇▇▇▇ 21%	sharp · at its level · small run · several runs
AIME 2025 lab +1	72.1% ✎	▁·▁▁▁▂▅▆█▇▅▁ beats 68/133	medium · avg 64%	▇▇▇▇▇▇▇▇▇ 21%	sharp · at its level · small run · several runs

other tests5 not in a task · 47% of its score

test	result	vs the others in its run	difficulty	counts	why
GPQA Diamond lab +1	53.4% ✎	▁▁▁▁▂▄▅▅███▄ beats 94/310	hard · avg 51%	▇▇▇▇▇▇▇▇▇ 22%	sharp · at its level · small run · several runs
Medmarks-V mean win rate (19 verifiable medical benchmarks, 71 configurations)	0.3741 ✎	▁▄▃▃▇▄█▃▅▁▁▁ beats 3/30	avg 0.423	▇▇▇▇▇▇ 15%	sharp
Medmarks-V weighted mean win rate (61 models, 71 configurations)	37.4% ✎	▂▃▅▂▆▃█▅█▅▃▆ beats 3/34	medium · avg 42%	▇▇▇ 7.2%	blunt · at its level
EuroEval HellaSwag (English), test split, 5-shot generative MC, accuracy	75.6% ✎	▃·▁▁▂▁▂▅█▇▄▁ beats 61/100	medium · avg 68%	▇ 1.7%	at its level · noisy
EuroEval Life in the UK (English knowledge MC), test split, 5-shot, accuracy	78.1% ✎	▁▁▁▁▁▁▁▂▄█▄▃ beats 60/146	easy · avg 83%	▇ 1.1%	noisy

$ where to run it

llama.cpp · recommended

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 ↗
Run Ministral 3 3B Reasoning 2512 · needs 2.15 GB · from huggingface.co/mistralai
chat in the terminal
llama cli -hf mistralai/Ministral-3-3B-Reasoning-2512-GGUF:Q4_K_M
Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf mistralai/Ministral-3-3B-Reasoning-2512-GGUF:Q4_K_M

LM Studio

version

Install LM Studio · Open the app once before using the lms command.
download it (macOS / Windows / Linux) ↗
Run Ministral 3 3B Reasoning 2512 · needs 2.15 GB · from huggingface.co/mistralai
open in LM Studio ↗
Or download it in the terminal
lms get https://huggingface.co/mistralai/Ministral-3-3B-Reasoning-2512-GGUF@Q4_K_M -y
Then run lms ls to see its name, and lms chat followed by that name

Ollama

Install Ollama
download it (macOS / Windows) ↗
Linux
curl -fsSL https://ollama.com/install.sh | sh
Run Ministral 3 3B Reasoning 2512 · needs 2.15 GB · from huggingface.co/mistralai
chat in the terminal
ollama run hf.co/mistralai/Ministral-3-3B-Reasoning-2512-GGUF:Q4_K_M

MLX (Mac)

version

Install MLX (Mac) · Apple silicon Macs only, text only.
with uv
uv tool install mlx-lm
or
pip install mlx-lm
Run Ministral 3 3B Reasoning 2512 · needs ~1.9 GB, Mac only · from huggingface.co/alexgusevski
chat in the terminal
mlx_lm.chat --model alexgusevski/Ministral-3-3B-Reasoning-2512-q4-mlx
Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model alexgusevski/Ministral-3-3B-Reasoning-2512-q4-mlx

Weights and documentation: huggingface.co/mistralai ↗

$ how the score was worked out · mistralai/Ministral-3-3B-Reas… 37%, mistralai/Ministral-3-8B-Reas… 37%, Medmarks: A Comprehensive Ope… 22% and 1 more

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

source	results	share of the score	licence
mistralai/Ministral-3-3B-Reasoning-2512 model card · lab-reported	4	▇▇▇▇▇▇▇▇▇▇▇ 37%	Apache-2.0
mistralai/Ministral-3-8B-Reasoning-2512 model card · lab-reported	4	▇▇▇▇▇▇▇▇▇▇▇ 37%	Apache-2.0
Medmarks: A Comprehensive Open-Source LLM Benchmark Suite for Medical Tasks — authors' own runs of 61 models	2	▇▇▇▇▇▇▇ 22%	CC BY 4.0
EuroEval leaderboards (results collection), English generative leaderboard	2	▇ 3%	MIT
prior · pull toward the middle		▇ 0%
weighted average		= 50.9

$ improve this page

Update model information Add a place to run it Add a test result Propose a benchmark Correct a benchmark score Edit 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 2026-10-11 02:20 UTC · sources and licences · json