What are the hardware requirements for Ministral 3 14B Reasoning 2512?

Ministral 3 14B Reasoning 2512 requires approximately 28 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Ministral 3 14B Reasoning 2512 and how large is it?

Ministral 3 14B Reasoning 2512 is developed by mistralai, a model with 14b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Ministral 3 14B Reasoning 2512?

Ministral 3 14B Reasoning 2512 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Ministral 3 14B Reasoning 2512 by mistralai — VRAM 28GB, 256K context

Name: Ministral 3 14B Reasoning 2512
Author: mistralai

Ministral 3 14B Reasoning 2512 is an open-source language model by mistralai. Features: 14b LLM, VRAM: 28GB, Context: 256K, License: MNPL-0.1, LLM Explorer Score: 0.28.

Arxiv:2601.08584 Ar Base model:finetune:mistralai/... Base model:mistralai/ministral... De En Es Fr It Ja Ko Mistral-common Mistral3 Nl Pt Region:us Safetensors Sharded Tensorflow Vllm Zh

Model Card on HF 🤗: https://huggingface.co/mistralai/Ministral-3-14B-Reasoning-2512

Ministral 3 14B Reasoning 2512 Benchmarks

LLME Score: 0.28363

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

Ministral 3 14B Reasoning 2512 (mistralai/Ministral-3-14B-Reasoning-2512)

Ministral 3 14B Reasoning 2512 Parameters and Internals

LLM Name	Ministral 3 14B Reasoning 2512
Repository 🤗	https://huggingface.co/mistralai/Ministral-3-14B-Reasoning-2512
Base Model(s)	Ministral 3 14B Base 2512 mistralai/Ministral-3-14B-Base-2512
Model Size	14b
Required VRAM	28 GB
Updated	2026-08-06
Maintainer	mistralai
Model Files	27.9 GB 4.9 GB: 1-of-6 4.9 GB: 2-of-6 4.9 GB: 3-of-6 4.9 GB: 4-of-6 4.9 GB: 5-of-6 3.5 GB: 6-of-6
Supported Languages	en fr es de it pt nl zh ja ko ar
Model Architecture	MambaSSM
License	MNPL-0.1
Context Length	262144
Model Max Length	262144
Tokenizer Class	LlamaTokenizerFast
Padding Token	<pad>
Vocabulary Size	131072

Quantized Models of the Ministral 3 14B Reasoning 2512

Model	Likes	Downloads	VRAM
... 14B Reasoning 2512 Uncensored	0	19	28 GB

Best Alternatives to Ministral 3 14B Reasoning 2512

Best Alternatives	Context / RAM	Downloads	Likes
Ministral 3 14B Instruct 2512	256K / 15.8 GB	110841	294
...stral 3 14B Instruct 2512 BF16	256K / 28 GB	18087	29
Ministral 3 14B Base 2512	256K / 28 GB	10597	60

Note: green Score (e.g. "73.2") means that the model is better than mistralai/Ministral-3-14B-Reasoning-2512.

Rank the Ministral 3 14B Reasoning 2512 Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

What open-source LLMs or SLMs are you in search of? 55617 in total.

Email us: info@extractum.io. Our Privacy Policy | Terms and Conditions | Suggest an improvement.

Our Social Media →

Original data from HuggingFace, Arena and various public git repos.

Check out Ag3ntum — our secure, self-hosted AI agent for server management.

Release v20260328a

Support LLM Explorer

Ministral 3 14B Reasoning 2512 by mistralai