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

SuperGLM 5.2 Abliterated NVFP4 by Jiunsong

 »  All LLMs  »  Jiunsong  »  SuperGLM 5.2 Abliterated NVFP4   URL Share it on

  2129   

SuperGLM 5.2 Abliterated NVFP4 is an open-source language model by Jiunsong. Features: 381b LLM, VRAM: 410GB, Context: 1024K, License: mit, MoE, LLM Explorer Score: 0.43.

  8-bit   Abliterated Base model:finetune:nvidia/glm... Base model:nvidia/glm-5.2-nvfp...   Conversational   Endpoints compatible   Glm   Glm-5.2   Glm moe dsa   Modelopt   Moe   Nvfp4   Obliteratus   Quantized   Region:us   Safetensors   Sglang   Sharded   Supertune   Tensorflow   Vllm
Model Card on HF 🤗: https://huggingface.co/Jiunsong/SuperGLM-5.2-abliterated-NVFP4 

SuperGLM 5.2 Abliterated NVFP4 Benchmarks

LLME Score: 0.42896
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
SuperGLM 5.2 Abliterated NVFP4 (Jiunsong/SuperGLM-5.2-abliterated-NVFP4)
🌟 Advertise your project 🚀

SuperGLM 5.2 Abliterated NVFP4 Parameters and Internals

LLM NameSuperGLM 5.2 Abliterated NVFP4
Repository 🤗https://huggingface.co/Jiunsong/SuperGLM-5.2-abliterated-NVFP4 
Base Model(s)  nvidia/GLM-5.2-NVFP4   nvidia/GLM-5.2-NVFP4
Model Size381b
Required VRAM410 GB
Updated2026-07-16
MaintainerJiunsong
Model Typeglm_moe_dsa
Model Files  10.0 GB: 1-of-47   10.0 GB: 2-of-47   10.0 GB: 3-of-47   10.0 GB: 4-of-47   10.0 GB: 5-of-47   10.0 GB: 6-of-47   10.0 GB: 7-of-47   10.0 GB: 8-of-47   10.0 GB: 9-of-47   10.0 GB: 10-of-47   10.0 GB: 11-of-47   10.0 GB: 12-of-47   10.0 GB: 13-of-47   10.0 GB: 14-of-47   10.0 GB: 15-of-47   10.0 GB: 16-of-47   10.0 GB: 17-of-47   10.0 GB: 18-of-47   10.0 GB: 19-of-47   10.0 GB: 20-of-47   10.0 GB: 21-of-47   10.0 GB: 22-of-47   10.0 GB: 23-of-47   10.0 GB: 24-of-47   10.0 GB: 25-of-47   10.0 GB: 26-of-47   10.0 GB: 27-of-47   10.0 GB: 28-of-47   10.0 GB: 29-of-47   10.0 GB: 30-of-47   10.0 GB: 31-of-47   10.0 GB: 32-of-47   10.0 GB: 33-of-47   10.0 GB: 34-of-47   10.0 GB: 35-of-47   10.0 GB: 36-of-47   10.0 GB: 37-of-47   10.0 GB: 38-of-47   10.0 GB: 39-of-47   10.0 GB: 40-of-47   10.0 GB: 41-of-47
Model ArchitectureGlmMoeDsaForCausalLM
Licensemit
Context Length1048576
Model Max Length1048576
Transformers Version5.11.0
Vocabulary Size154880

Best Alternatives to SuperGLM 5.2 Abliterated NVFP4

Best Alternatives
Context / RAM
Downloads
Likes
GLM 5.2 NVFP41024K / 429.9 GB832851260
GLM 5.2 NVFP41024K / 429.9 GB3333
Note: green Score (e.g. "73.2") means that the model is better than Jiunsong/SuperGLM-5.2-abliterated-NVFP4.

Rank the SuperGLM 5.2 Abliterated NVFP4 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? 55138 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