What are the hardware requirements for Magistral Small 2506 Bnb 4bit?

Magistral Small 2506 Bnb 4bit requires approximately 14.2 GB of VRAM and supports a context window of 40K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Magistral Small 2506 Bnb 4bit and how large is it?

Magistral Small 2506 Bnb 4bit is developed by unsloth, a model with 24b 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 Magistral Small 2506 Bnb 4bit?

Magistral Small 2506 Bnb 4bit is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Magistral Small 2506 Bnb 4bit by unsloth — VRAM 14.2GB, 40K context

Name: Magistral Small 2506 Bnb 4bit
Author: unsloth

Magistral Small 2506 Bnb 4bit is an open-source language model by unsloth. Features: 24b LLM, VRAM: 14.2GB, Context: 40K, License: apache-2.0, Quantized, Instruction-Based, LLM Explorer Score: 0.19.

4-bit 4bit Ar Base model:mistralai/magistral... Base model:quantized:mistralai... Bitsandbytes Bn Conversational De En Es Fa Fr Hi Id Instruct It Ja Ko Mistral Ms Ne Pl Pt Quantized Region:us Ro Ru Safetensors Sharded Sr Sv Tensorflow Tr Uk Unsloth Vi Zh

Model Card on HF 🤗: https://huggingface.co/unsloth/Magistral-Small-2506-bnb-4bit

Magistral Small 2506 Bnb 4bit Parameters and Internals

LLM Name	Magistral Small 2506 Bnb 4bit
Repository 🤗	https://huggingface.co/unsloth/Magistral-Small-2506-bnb-4bit
Base Model(s)	Magistral Small 2506 mistralai/Mistral-Small-3.1-24B-Instruct-2503 mistralai/Magistral-Small-2506 mistralai/Mistral-Small-3.1-24B-Instruct-2503
Model Size	24b
Required VRAM	14.2 GB
Updated	2026-08-10
Maintainer	unsloth
Model Type	mistral
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-3 5.0 GB: 2-of-3 4.2 GB: 3-of-3
Supported Languages	en fr de es pt it ja ko ru zh ar fa id ms ne pl ro sr sv tr uk vi hi bn
Quantization Type	4bit
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	40960
Model Max Length	40960
Transformers Version	4.52.4
Tokenizer Class	LlamaTokenizerFast
Padding Token	<pad>
Vocabulary Size	131072
Torch Data Type	bfloat16

Rank the Magistral Small 2506 Bnb 4bit capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

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

Best Alternatives to Magistral Small 2506 Bnb 4bit

Best Alternatives	Context / RAM	Downloads	Likes
Devstral Small 2507 Bnb 4bit	128K / 14.2 GB	398	1
...al Small 2507 Unsloth Bnb 4bit	128K / 15 GB	128	5
...al Small 2506 Unsloth Bnb 4bit	40K / 15 GB	109	4
...l Small 24B Instruct 2501 4bit	32K / 13.3 GB	1344	14
Not So Small Alpaca 24B	32K / 47.3 GB	14	0
Napoleon 24B V0.0	32K / 47.3 GB	18	0
...all 24B Instruct 2501 Bnb 4bit	32K / 14.2 GB	4725	3
...Instruct 2501 Unsloth Bnb 4bit	32K / 15 GB	2955	19
...b With Nemotron Converted Vllm	32K / 47.3 GB	38	0
...24B Base 2501 Unsloth Bnb 4bit	32K / 15.7 GB	116	3

Note: green Score (e.g. "73.2") means that the model is better than unsloth/Magistral-Small-2506-bnb-4bit.