What are the hardware requirements for MS3.2 24B Astral Mirage?

MS3.2 24B Astral Mirage requires approximately 47.3 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed MS3.2 24B Astral Mirage and how large is it?

MS3.2 24B Astral Mirage is developed by Vortex5, 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 MS3.2 24B Astral Mirage?

MS3.2 24B Astral Mirage is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

MS3.2 24B Astral Mirage by Vortex5 — VRAM 47.3GB, 128K context

Name: MS3.2 24B Astral Mirage
Author: Vortex5

MS3.2 24B Astral Mirage is an open-source language model by Vortex5. Features: 24b LLM, VRAM: 47.3GB, Context: 128K, Merged, LLM Explorer Score: 0.19.

Merged Model Arxiv:2403.19522 Base model:doctor-shotgun/ms3.... Base model:gryphe/codex-24b-sm... Base model:zerofata/ms3.2-pain... Endpoints compatible Mistral Region:us Roleplay Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Vortex5/MS3.2-24B-Astral-Mirage

MS3.2 24B Astral Mirage Parameters and Internals

LLM Name	MS3.2 24B Astral Mirage
Repository 🤗	https://huggingface.co/Vortex5/MS3.2-24B-Astral-Mirage
Base Model(s)	Doctor-Shotgun/MS3.2-24B-Magnum-Diamond zerofata/MS3.2-PaintedFantasy-24B Gryphe/Codex-24B-Small-3.2 Doctor-Shotgun/MS3.2-24B-Magnum-Diamond zerofata/MS3.2-PaintedFantasy-24B Gryphe/Codex-24B-Small-3.2
Merged Model	Yes
Model Size	24b
Required VRAM	47.3 GB
Updated	2026-07-12
Maintainer	Vortex5
Model Type	mistral
Model Files	4.9 GB: 1-of-10 4.8 GB: 2-of-10 4.8 GB: 3-of-10 4.9 GB: 4-of-10 4.8 GB: 5-of-10 4.8 GB: 6-of-10 4.9 GB: 7-of-10 4.8 GB: 8-of-10 4.8 GB: 9-of-10 3.8 GB: 10-of-10
Model Architecture	MistralForCausalLM
Context Length	131072
Model Max Length	131072
Transformers Version	4.53.0
Tokenizer Class	LlamaTokenizerFast
Padding Token	<pad>
Vocabulary Size	131072
Torch Data Type	bfloat16

Rank the MS3.2 24B Astral Mirage 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 MS3.2 24B Astral Mirage

Best Alternatives	Context / RAM	Downloads	Likes
...ral Nemo Instruct 24B Merge V1	1000K / 24.6 GB	9	0
...all 3.1 24B Instruct 2503 Text	128K / 47.3 GB	16	0
Morbid Miasma 24B	128K / 47.3 GB	33	9
MagiSeek Pro V1	128K / 47.3 GB	303	1
Slimaki Tavern 24B V1.3	128K / 47.3 GB	346	16
Zepar 24B V1	128K / 47.3 GB	36	3
Cydonia 24B V4.3	128K / 47.3 GB	3652	138
Transformed Journey 24B	128K / 47.3 GB	65	0
Magistry Painted Cydonia 24B	128K / 47.3 GB	60	2
Morax 24B V2	128K / 18.3 GB	57	2

Note: green Score (e.g. "73.2") means that the model is better than Vortex5/MS3.2-24B-Astral-Mirage.