What are the hardware requirements for Ministral 3 8B Instruct 2512?

Ministral 3 8B Instruct 2512 requires approximately 10.4 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 8B Instruct 2512 and how large is it?

Ministral 3 8B Instruct 2512 is developed by mistralai, a model with 8b 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 8B Instruct 2512?

Ministral 3 8B Instruct 2512 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Ministral 3 8B Instruct 2512 by mistralai — VRAM 10.4GB, 256K context

Name: Ministral 3 8B Instruct 2512
Author: mistralai

Ministral 3 8B Instruct 2512 is an open-source language model by mistralai. Features: 8b LLM, VRAM: 10.4GB, Context: 256K, License: MNPL-0.1, Instruction-Based, LLM Explorer Score: 0.32.

Arxiv:2601.08584 Ar Base model:mistralai/ministral... Base model:quantized:mistralai... De En Es Fp8 Fr Instruct It Ja Ko Mistral-common Mistral3 Nl Pt Region:us Safetensors Sharded Tensorflow Vllm Zh

Model Card on HF 🤗: https://huggingface.co/mistralai/Ministral-3-8B-Instruct-2512

Ministral 3 8B Instruct 2512 Benchmarks

LLME Score: 0.31624

^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 8B Instruct 2512 (mistralai/Ministral-3-8B-Instruct-2512)

Ministral 3 8B Instruct 2512 Parameters and Internals

LLM Name	Ministral 3 8B Instruct 2512
Repository 🤗	https://huggingface.co/mistralai/Ministral-3-8B-Instruct-2512
Base Model(s)	Ministral 3 8B Base 2512 mistralai/Ministral-3-8B-Base-2512
Model Size	8b
Required VRAM	10.4 GB
Updated	2026-07-29
Maintainer	mistralai
Instruction-Based	Yes
Model Files	10.4 GB 5.0 GB: 1-of-3 5.0 GB: 2-of-3 0.4 GB: 3-of-3
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	TokenizersBackend
Padding Token	<pad>
Vocabulary Size	131072

Quantized Models of the Ministral 3 8B Instruct 2512

Model	Likes	Downloads	VRAM
Amaretto 8B	0	442	17 GB
...Instruct 2512 Unsloth Bnb 4bit	1	2580	7 GB
...al 3 8B Instruct 2512 Bnb 4bit	0	255	6 GB

Best Alternatives to Ministral 3 8B Instruct 2512

Best Alternatives	Context / RAM	Downloads	Likes
...istral 3 8B Instruct 2512 BF16	256K / 17.8 GB	16213	23
...8B Instruct 2512 BF16 SOM MPOA	256K / 17.8 GB	50	0
Ministral 8B Instruct 2410	0K / 16.1 GB	299686	584
Ministral 8B Instruct 2410 HF	0K / 16.1 GB	7	0

Note: green Score (e.g. "73.2") means that the model is better than mistralai/Ministral-3-8B-Instruct-2512.

Rank the Ministral 3 8B Instruct 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? 55501 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 8B Instruct 2512 by mistralai