What are the hardware requirements for Devstral 2 123B Instruct 2512 6bit?

Devstral 2 123B Instruct 2512 6bit requires approximately 101.2 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 Devstral 2 123B Instruct 2512 6bit and how large is it?

Devstral 2 123B Instruct 2512 6bit is developed by mlx-community, a model with 123b 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 Devstral 2 123B Instruct 2512 6bit?

Devstral 2 123B Instruct 2512 6bit is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Devstral 2 123B Instruct 2512 6bit by mlx-community — VRAM 101.2GB, 256K context

Name: Devstral 2 123B Instruct 2512 6bit
Author: mlx-community

Devstral 2 123B Instruct 2512 6bit is an open-source language model by mlx-community. Features: 123b LLM, VRAM: 101.2GB, Context: 256K, License: other, Quantized, Instruction-Based, LLM Explorer Score: 0.21.

6-bit 6bit Base model:mistralai/devstral-... Base model:quantized:mistralai... Conversational Instruct Ministral3 Mistral-common Mlx Quantized Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/mlx-community/Devstral-2-123B-Instruct-2512-6bit

Devstral 2 123B Instruct 2512 6bit Parameters and Internals

LLM Name	Devstral 2 123B Instruct 2512 6bit
Repository 🤗	https://huggingface.co/mlx-community/Devstral-2-123B-Instruct-2512-6bit
Base Model(s)	Devstral 2 123B Instruct 2512 mistralai/Devstral-2-123B-Instruct-2512
Model Size	123b
Required VRAM	101.2 GB
Updated	2026-09-18
Maintainer	mlx-community
Model Type	ministral3
Instruction-Based	Yes
Model Files	5.2 GB: 1-of-20 5.3 GB: 2-of-20 5.4 GB: 3-of-20 5.3 GB: 4-of-20 5.3 GB: 5-of-20 5.3 GB: 6-of-20 5.4 GB: 7-of-20 5.3 GB: 8-of-20 5.3 GB: 9-of-20 5.3 GB: 10-of-20 5.4 GB: 11-of-20 5.3 GB: 12-of-20 5.3 GB: 13-of-20 5.3 GB: 14-of-20 5.4 GB: 15-of-20 5.3 GB: 16-of-20 5.3 GB: 17-of-20 5.3 GB: 18-of-20 4.2 GB: 19-of-20 1.3 GB: 20-of-20
Quantization Type	6bit
Model Architecture	Ministral3ForCausalLM
License	other
Context Length	262144
Model Max Length	262144
Transformers Version	5.0.0.dev0
Tokenizer Class	TokenizersBackend
Padding Token	<pad>
Vocabulary Size	131072

Rank the Devstral 2 123B Instruct 2512 6bit 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 Devstral 2 123B Instruct 2512 6bit

Best Alternatives	Context / RAM	Downloads	Likes
...tral 2 123B Instruct 2512 4bit	256K / 70 GB	230	1
...tral 2 123B Instruct 2512 8bit	256K / 133 GB	116	1
Devstral 2 123B Instruct 2512	256K / 196.3 GB	57919	329
Devstral 2 123B Instruct 2512	256K / 69.6 GB	43	0

Note: green Score (e.g. "73.2") means that the model is better than mlx-community/Devstral-2-123B-Instruct-2512-6bit.