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

Devstral 2 123B Instruct 2512 requires approximately 196.3 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 and how large is it?

Devstral 2 123B Instruct 2512 is developed by mistralai, 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?

Devstral 2 123B Instruct 2512 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 by mistralai — VRAM 196.3GB, 256K context

Name: Devstral 2 123B Instruct 2512
Author: mistralai

Devstral 2 123B Instruct 2512 is an open-source language model by mistralai. Features: 123b LLM, VRAM: 196.3GB, Context: 256K, License: other, Instruction-Based, LLM Explorer Score: 0.31.

Fp8 Instruct Ministral3 Mistral-common Region:us Safetensors Sharded Tensorflow Vllm

Model Card on HF 🤗: https://huggingface.co/mistralai/Devstral-2-123B-Instruct-2512

Devstral 2 123B Instruct 2512 Parameters and Internals

LLM Name	Devstral 2 123B Instruct 2512
Repository 🤗	https://huggingface.co/mistralai/Devstral-2-123B-Instruct-2512
Model Size	123b
Required VRAM	196.3 GB
Updated	2026-08-02
Maintainer	mistralai
Model Type	ministral3
Instruction-Based	Yes
Model Files	4.8 GB: 1-of-27 4.9 GB: 2-of-27 4.8 GB: 3-of-27 4.9 GB: 4-of-27 4.8 GB: 5-of-27 4.9 GB: 6-of-27 4.8 GB: 7-of-27 4.9 GB: 8-of-27 4.8 GB: 9-of-27 4.9 GB: 10-of-27 4.8 GB: 11-of-27 4.9 GB: 12-of-27 4.8 GB: 13-of-27 4.9 GB: 14-of-27 4.8 GB: 15-of-27 4.9 GB: 16-of-27 4.8 GB: 17-of-27 4.9 GB: 18-of-27 4.8 GB: 19-of-27 4.9 GB: 20-of-27 4.8 GB: 21-of-27 4.9 GB: 22-of-27 4.8 GB: 23-of-27 4.9 GB: 24-of-27 4.8 GB: 25-of-27 3.9 GB: 26-of-27 3.2 GB: 27-of-27 4.9 GB: 1-of-27 4.9 GB: 2-of-27 4.8 GB: 3-of-27 4.9 GB: 4-of-27 4.8 GB: 5-of-27 4.9 GB: 6-of-27 4.8 GB: 7-of-27 4.9 GB: 8-of-27 4.8 GB: 9-of-27 4.9 GB: 10-of-27 4.8 GB: 11-of-27 4.9 GB: 12-of-27 4.8 GB: 13-of-27 4.9 GB: 14-of-27
Model Architecture	Ministral3ForCausalLM
License	other
Context Length	262144
Model Max Length	262144
Transformers Version	5.0.0.dev0
Vocabulary Size	131072

Rank the Devstral 2 123B Instruct 2512 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

Quantized Models of the Devstral 2 123B Instruct 2512

Model	Likes	Downloads	VRAM
...tral 2 123B Instruct 2512 4bit	1	230	70 GB
...tral 2 123B Instruct 2512 8bit	1	116	133 GB

Best Alternatives to Devstral 2 123B Instruct 2512

Best Alternatives	Context / RAM	Downloads	Likes
Devstral 2 123B Instruct 2512	256K / 69.6 GB	30	0
...tral 2 123B Instruct 2512 4bit	256K / 70 GB	230	1
...tral 2 123B Instruct 2512 8bit	256K / 133 GB	116	1