LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  Agents   Benchmarks   LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

SuperHY3 Abliterated MLX 4bit by Jiunsong

 »  All LLMs  »  Jiunsong  »  SuperHY3 Abliterated MLX 4bit   URL Share it on

  193   

SuperHY3 Abliterated MLX 4bit is an open-source language model by Jiunsong. Features: 48.1b LLM, VRAM: 169.9GB, Context: 256K, License: apache-2.0, Quantized, LLM Explorer Score: 0.34.

  4-bit   4bit   Abliterated   Apple-silicon Base model:jiunsong/superhy3-a... Base model:quantized:jiunsong/...   Conversational   Hunyuan   Hy3   Hy v3   Imatrix   Mlx   Moe   Obliteratus   Quantized   Region:us   Safetensors   Sharded   Supertune   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Jiunsong/SuperHY3-abliterated-MLX-4bit 

SuperHY3 Abliterated MLX 4bit Benchmarks

LLME Score: 0.33774
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
SuperHY3 Abliterated MLX 4bit (Jiunsong/SuperHY3-abliterated-MLX-4bit)
🌟 Advertise your project 🚀

SuperHY3 Abliterated MLX 4bit Parameters and Internals

LLM NameSuperHY3 Abliterated MLX 4bit
Repository 🤗https://huggingface.co/Jiunsong/SuperHY3-abliterated-MLX-4bit 
Base Model(s)  SuperHY3 Abliterated NVFP4   Jiunsong/SuperHY3-abliterated-NVFP4
Model Size48.1b
Required VRAM169.9 GB
Updated2026-07-14
MaintainerJiunsong
Model Typehy_v3
Model Files  6.0 GB: 1-of-32   5.3 GB: 2-of-32   5.4 GB: 3-of-32   5.4 GB: 4-of-32   5.4 GB: 5-of-32   5.4 GB: 6-of-32   5.4 GB: 7-of-32   5.4 GB: 8-of-32   5.4 GB: 9-of-32   5.4 GB: 10-of-32   5.4 GB: 11-of-32   5.4 GB: 12-of-32   5.4 GB: 13-of-32   5.4 GB: 14-of-32   5.4 GB: 15-of-32   5.4 GB: 16-of-32   5.4 GB: 17-of-32   5.4 GB: 18-of-32   5.4 GB: 19-of-32   5.4 GB: 20-of-32   5.4 GB: 21-of-32   5.4 GB: 22-of-32   5.4 GB: 23-of-32   5.4 GB: 24-of-32   5.4 GB: 25-of-32   5.4 GB: 26-of-32   5.4 GB: 27-of-32   5.4 GB: 28-of-32   5.4 GB: 29-of-32   5.4 GB: 30-of-32   5.4 GB: 31-of-32   2.0 GB: 32-of-32
Quantization Type4bit
Model ArchitectureHYV3ForCausalLM
Licenseapache-2.0
Context Length262144
Model Max Length262144
Transformers Version5.6.0
Vocabulary Size120832

Rank the SuperHY3 Abliterated MLX 4bit Capabilities

🆘 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! 🌟

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? 55092 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer