LLM EXPLORER 67,808 MODELS INDEXED

Gemma 4 E4B Abliterated by DuoNeural

By DuoNeural · 33 downloads

Gemma 4 E4B Abliterated is an open-source language model by DuoNeural. Features: 8b LLM, VRAM: 16GB, LLM Explorer Score: 0.22.

  Gemma4   Region:us   Safetensors   Sharded   Tensorflow

Gemma 4 E4B Abliterated Parameters and Internals

LLM NameGemma 4 E4B Abliterated
Repository πŸ€—https://huggingface.co/DuoNeural/Gemma-4-E4B-Abliterated 
Model Size8b
Required VRAM16 GB
Updated2026-11-18
MaintainerDuoNeural
Model Typegemma4
Model Files  5.6 GB: 1-of-4   5.0 GB: 2-of-4   5.0 GB: 3-of-4   0.4 GB: 4-of-4   0.0 GB
Model ArchitectureGemma4ForConditionalGeneration
Transformers Version5.5.0
Tokenizer ClassGemmaTokenizer
Padding Token<pad>

Best Alternatives to Gemma 4 E4B Abliterated

Best Alternatives
Context / RAM
Downloads
Likes
Gemma 4 E4B It0K / 16 GB40278071785
Gemma 4 E4B It0K / 16 GB22990823
Gemma 4 E4B0K / 16 GB268109386
...DIA Nemotron 3.5 Bolt E4B BF160K / 15.9 GB341856
...4B It Weight Space Programming0K / 16 GB653
...otron 3.5 Bolt E4B Artemis FP80K / 2 GB613
Vishwamai Stage10K / 16 GB4720
...emma4 E4B ASR PDEFAULT E2 BF160K / 16 GB330
...Nemotron 3.5 Bolt E4B MLX OQ8e0K / 8.9 GB1501
... E4B FACEBOOK PDEFAULT E2 BF160K / 16 GB330
Note: green Score (e.g. "73.2") means that the model is better than DuoNeural/Gemma-4-E4B-Abliterated.