LLM EXPLORER 59,358 MODELS INDEXED

Aura MoE 2x4B by AuraIndustries

By AuraIndustries · 8 downloads

Aura MoE 2x4B is an open-source language model by AuraIndustries. Features: 7.2b LLM, VRAM: 14.5GB, Context: 128K, License: apache-2.0, MoE, Instruction-Based, LLM Explorer Score: 0.2.

Base model:finetune:interviten... Base model:intervitensinc/llam... Dataset:anthracite-core/full-o... Dataset:fourohfour/instruct ph...   Dataset:fourohfour/rp phase   Dataset:mielikki/erebus-87k   En   Instruct   Mixtral   Moe   Region:us   Safetensors   Sharded   Tensorflow

Aura MoE 2x4B Parameters and Internals

LLM NameAura MoE 2x4B
Repository πŸ€—https://huggingface.co/AuraIndustries/Aura-MoE-2x4B 
Base Model(s)  IntervitensInc/Llama-3.1-Minitron-4B-Width-Base-chatml   IntervitensInc/Llama-3.1-Minitron-4B-Width-Base-chatml
Model Size7.2b
Required VRAM14.5 GB
Updated2025-03-18
MaintainerAuraIndustries
Model Typemixtral
Instruction-BasedYes
Model Files  5.0 GB: 1-of-3   5.0 GB: 2-of-3   4.5 GB: 3-of-3
Supported Languagesen
Model ArchitectureMixtralForCausalLM
Licenseapache-2.0
Context Length131072
Model Max Length131072
Transformers Version4.47.0
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|finetune_right_pad_id|>
Vocabulary Size128256
Torch Data Typebfloat16