LLM EXPLORER 61,918 MODELS INDEXED

Goetia 26B A4B V1.3 Absolute Heretic ARA Mlx 8Bit by otherwhere1

By otherwhere1 · 125 downloads
HF Repo ↗ Copy URL

Goetia 26B A4B V1.3 Absolute Heretic ARA Mlx 8Bit is an open-source language model by otherwhere1. Features: 25.2b LLM, VRAM: 26.8GB, License: apache-2.0, MoE, Quantized, LLM Explorer Score: 0.31.

  8-bit   8bit   Ablated   Ara   Ari Base model:naphula/goetia-26b-... Base model:quantized:naphula/g...   Conversational   Creative writing   Endpoints compatible   Eng   Gemma4   Heretic   Image-text-to-text   Merge   Mergekit   Mlx   Mlx-my-repo   Moe   Moe della   Quantized   Region:us   Roleplay   Safetensors   Sharded   Tensorflow   Uncensored
Model Card on HF 🤗: https://huggingface.co/otherwhere1/Goetia-26B-A4B-v1.3-Absolute-Heretic-ARA-mlx-8Bit 

Goetia 26B A4B V1.3 Absolute Heretic ARA Mlx 8Bit Parameters and Internals

LLM NameGoetia 26B A4B V1.3 Absolute Heretic ARA Mlx 8Bit
Repository 🤗https://huggingface.co/otherwhere1/Goetia-26B-A4B-v1.3-Absolute-Heretic-ARA-mlx-8Bit 
Base Model(s)  Naphula/Goetia-26B-A4B-v1.3-Absolute-Heretic-ARA   Naphula/Goetia-26B-A4B-v1.3-Absolute-Heretic-ARA
Model Size25.2b
Required VRAM26.8 GB
Updated2026-09-22
Maintainerotherwhere1
Model Typegemma4
Model Files  5.2 GB: 1-of-6   5.2 GB: 2-of-6   5.2 GB: 3-of-6   5.2 GB: 4-of-6   5.2 GB: 5-of-6   0.8 GB: 6-of-6
Supported Languagesen
Quantization Type8bit
Model ArchitectureGemma4ForConditionalGeneration
Licenseapache-2.0
Transformers Version5.12.1
Tokenizer ClassGemmaTokenizer
Padding Token<pad>
Vocabulary Size262144

Best Alternatives to Goetia 26B A4B V1.3 Absolute Heretic ARA Mlx 8Bit

Best Alternatives
Context / RAM
Downloads
Likes
...a 26B A4B V1.4 LazyLora Heresy0K / 50.5 GB227