What are the hardware requirements for Mistral 2x22B MoE Power Codestral Ultimate 39B?

Mistral 2x22B MoE Power Codestral Ultimate 39B requires approximately 78.3 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Mistral 2x22B MoE Power Codestral Ultimate 39B and how large is it?

Mistral 2x22B MoE Power Codestral Ultimate 39B is developed by DavidAU, a model with 39.2b 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 Mistral 2x22B MoE Power Codestral Ultimate 39B?

Mistral 2x22B MoE Power Codestral Ultimate 39B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Mistral 2x22B MoE Power Codestral Ultimate 39B by DavidAU — VRAM 78.3GB, 32K context

Name: Mistral 2x22B MoE Power Codestral Ultimate 39B
Author: DavidAU

Mistral 2x22B MoE Power Codestral Ultimate 39B is an open-source language model by DavidAU. Features: 39.2b LLM, VRAM: 78.3GB, Context: 32K, License: MNPL-0.1, MoE, LLM Explorer Score: 0.19.

2x22b Autotrain compatible Base model:merge:migtissera/tr... Base model:merge:mistralai/cod... Base model:migtissera/trinity-... Base model:mistralai/codestral... Chat Code Code generation Codegen Coder Codestral Coding Conversational En Endpoints compatible Merge Mistral Mistral moe Mixtral Mixture of experts Moe Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/DavidAU/Mistral-2x22B-MOE-Power-Codestral-Ultimate-39B

Mistral 2x22B MoE Power Codestral Ultimate 39B Parameters and Internals

LLM Name	Mistral 2x22B MoE Power Codestral Ultimate 39B
Repository 🤗	https://huggingface.co/DavidAU/Mistral-2x22B-MOE-Power-Codestral-Ultimate-39B
Base Model(s)	mistralai/Codestral-22B-v0.1 migtissera/Trinity-2-Codestral-22B-v0.2 mistralai/Codestral-22B-v0.1 migtissera/Trinity-2-Codestral-22B-v0.2
Model Size	39.2b
Required VRAM	78.3 GB
Updated	2025-10-05
Maintainer	DavidAU
Model Type	mixtral
Model Files	4.9 GB: 1-of-16 5.0 GB: 2-of-16 4.9 GB: 3-of-16 5.0 GB: 4-of-16 4.9 GB: 5-of-16 4.9 GB: 6-of-16 5.0 GB: 7-of-16 4.9 GB: 8-of-16 5.0 GB: 9-of-16 4.9 GB: 10-of-16 5.0 GB: 11-of-16 4.9 GB: 12-of-16 4.9 GB: 13-of-16 5.0 GB: 14-of-16 4.9 GB: 15-of-16 4.2 GB: 16-of-16
Supported Languages	en
Model Architecture	MixtralForCausalLM
License	MNPL-0.1
Context Length	32768
Model Max Length	32768
Transformers Version	4.52.0.dev0
Tokenizer Class	LlamaTokenizer
Padding Token	<s>
Vocabulary Size	32768
Torch Data Type	bfloat16

Rank the Mistral 2x22B MoE Power Codestral Ultimate 39B 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