Mistral 2x24B MoE Power Magistral Devstral Reasoning Ultimate 44B by DavidAU

 ยป  All LLMs  ยป  DavidAU  ยป  Mistral 2x24B MoE Power Magistral Devstral Reasoning Ultimate 44B   URL Share it on

  Arxiv:2506.10910   2x24b   Ar   Autotrain compatible Base model:merge:mistralai/dev... Base model:merge:mistralai/mag... Base model:mistralai/devstral-... Base model:mistralai/magistral...   Bn   Chat   Code   Code generation   Codegen   Coder   Codestral   Coding   Conversational   De   Devstral   En   Endpoints compatible   Es   Fa   Fr   Hi   Id   It   Ja   Ko   Magistral   Merge   Mistral   Mistral moe   Mixtral   Mixture of experts   Moe   Ms   Ne   Pl   Pt   Reasoning   Region:us   Ro   Ru   Safetensors   Sharded   Sr   Sv   Tensorflow   Thinking   Tr   Uk   Vi   Zh

Mistral 2x24B MoE Power Magistral Devstral Reasoning Ultimate 44B Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Mistral 2x24B MoE Power Magistral Devstral Reasoning Ultimate 44B (DavidAU/Mistral-2x24B-MOE-Power-Magistral-Devstral-Reasoning-Ultimate-44B)
๐ŸŒŸ Advertise your project ๐Ÿš€

Mistral 2x24B MoE Power Magistral Devstral Reasoning Ultimate 44B Parameters and Internals

LLM NameMistral 2x24B MoE Power Magistral Devstral Reasoning Ultimate 44B
Repository ๐Ÿค—https://huggingface.co/DavidAU/Mistral-2x24B-MOE-Power-Magistral-Devstral-Reasoning-Ultimate-44B 
Base Model(s)  Devstral Small 2505   Magistral Small 2506   mistralai/Devstral-Small-2505   mistralai/Magistral-Small-2506
Model Size43.7b
Required VRAM87.8 GB
Updated2025-07-19
MaintainerDavidAU
Model Typemixtral
Model Files  4.9 GB: 1-of-18   5.0 GB: 2-of-18   5.0 GB: 3-of-18   4.9 GB: 4-of-18   5.0 GB: 5-of-18   5.0 GB: 6-of-18   4.9 GB: 7-of-18   5.0 GB: 8-of-18   5.0 GB: 9-of-18   4.9 GB: 10-of-18   5.0 GB: 11-of-18   5.0 GB: 12-of-18   4.9 GB: 13-of-18   5.0 GB: 14-of-18   5.0 GB: 15-of-18   4.9 GB: 16-of-18   5.0 GB: 17-of-18   3.4 GB: 18-of-18
Supported Languagesen fr de es pt it ja ko ru zh ar fa id ms ne pl ro sr sv tr uk vi hi bn
Model ArchitectureMixtralForCausalLM
Licenseapache-2.0
Context Length131072
Model Max Length131072
Transformers Version4.52.0.dev0
Tokenizer ClassLlamaTokenizerFast
Padding Token<s>
Vocabulary Size131072
Torch Data Typebfloat16

Rank the Mistral 2x24B MoE Power Magistral Devstral Reasoning Ultimate 44B Capabilities

๐Ÿ†˜ Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! ๐ŸŒŸ

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  

What open-source LLMs or SLMs are you in search of? 49901 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124