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

DeepSeek V4 Flash 0731 Abliterated FP8 by apetersson

 »  All LLMs  »  apetersson  »  DeepSeek V4 Flash 0731 Abliterated FP8   URL Share it on

  4502   

DeepSeek V4 Flash 0731 Abliterated FP8 is an open-source language model by apetersson. Features: 304.2b LLM, VRAM: 145.1GB, Context: 1024K, License: mit, LLM Explorer Score: 0.43.

  8-bit   Abliterated Base model:deepseek-ai/deepsee... Base model:finetune:deepseek-a...   Deepseek-v4   Deepseek v4   Endpoints compatible   Fp8   Mechanistic-interpretability   Mixture-of-experts   Model-surgery   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/apetersson/DeepSeek-V4-Flash-0731-Abliterated-FP8 

DeepSeek V4 Flash 0731 Abliterated FP8 Benchmarks

LLME Score: 0.4295
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
DeepSeek V4 Flash 0731 Abliterated FP8 (apetersson/DeepSeek-V4-Flash-0731-Abliterated-FP8)

DeepSeek V4 Flash 0731 Abliterated FP8 Parameters and Internals

LLM NameDeepSeek V4 Flash 0731 Abliterated FP8
Repository 🤗https://huggingface.co/apetersson/DeepSeek-V4-Flash-0731-Abliterated-FP8 
Base Model(s)  deepseek-ai/DeepSeek-V4-Flash-0731   deepseek-ai/DeepSeek-V4-Flash-0731
Model Size304.2b
Required VRAM145.1 GB
Updated2026-08-04
Maintainerapetersson
Model Typedeepseek_v4
Model Files  1.1 GB: 1-of-48   3.6 GB: 2-of-48   3.6 GB: 3-of-48   3.6 GB: 4-of-48   3.6 GB: 5-of-48   3.6 GB: 6-of-48   3.6 GB: 7-of-48   3.6 GB: 8-of-48   3.6 GB: 9-of-48   3.6 GB: 10-of-48   3.6 GB: 11-of-48   3.6 GB: 12-of-48   3.6 GB: 13-of-48   3.6 GB: 14-of-48   3.6 GB: 15-of-48   3.6 GB: 16-of-48   3.6 GB: 17-of-48   3.6 GB: 18-of-48   3.6 GB: 19-of-48   3.6 GB: 20-of-48   3.6 GB: 21-of-48   3.6 GB: 22-of-48   3.6 GB: 23-of-48   3.6 GB: 24-of-48   3.6 GB: 25-of-48   3.6 GB: 26-of-48   3.6 GB: 27-of-48   3.6 GB: 28-of-48   3.6 GB: 29-of-48   3.6 GB: 30-of-48   3.6 GB: 31-of-48   3.6 GB: 32-of-48   3.6 GB: 33-of-48   3.6 GB: 34-of-48   3.6 GB: 35-of-48   3.6 GB: 36-of-48   3.6 GB: 37-of-48   3.6 GB: 38-of-48   3.6 GB: 39-of-48   3.6 GB: 40-of-48   3.6 GB: 41-of-48
Model ArchitectureDeepseekV4ForCausalLM
Licensemit
Context Length1048576
Model Max Length1048576
Transformers Version4.57.1
Vocabulary Size129280
Torch Data Typebfloat16

Best Alternatives to DeepSeek V4 Flash 0731 Abliterated FP8

Best Alternatives
Context / RAM
Downloads
Likes
DeepSeek V4 Flash 0731 NVFP41024K / 149.3 GB74216
DeepSeek V4 Flash 07311024K / 148.7 GB8925
DeepSeek V4 Flash 0731 NVFP41024K / 156.9 GB752

Rank the DeepSeek V4 Flash 0731 Abliterated FP8 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? 55613 in total.

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