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

DeepSeek V4 Flash 4bit by mlx-community

 »  All LLMs  »  mlx-community  »  DeepSeek V4 Flash 4bit   URL Share it on

  10358   

DeepSeek V4 Flash 4bit is an open-source language model by mlx-community. Features: 284.3b LLM, VRAM: 152.2GB, Context: 1024K, Quantized, LLM Explorer Score: 0.47, ELO: 1433.

  4-bit   4bit   Conversational   Deepseek v4   En   Mlx   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/mlx-community/DeepSeek-V4-Flash-4bit 

DeepSeek V4 Flash 4bit Benchmarks

LMSys ELO: 1433 vs 1272 (so35)12.7%
LLME Score: 0.47424
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 4bit (mlx-community/DeepSeek-V4-Flash-4bit)
🌟 Advertise your project 🚀

DeepSeek V4 Flash 4bit Parameters and Internals

LLM NameDeepSeek V4 Flash 4bit
Repository 🤗https://huggingface.co/mlx-community/DeepSeek-V4-Flash-4bit 
Model Size284.3b
Required VRAM152.2 GB
Updated2026-06-17
Maintainermlx-community
Model Typedeepseek_v4
Model Files  5.0 GB: 1-of-33   4.7 GB: 2-of-33   4.7 GB: 3-of-33   4.7 GB: 4-of-33   4.7 GB: 5-of-33   4.7 GB: 6-of-33   4.7 GB: 7-of-33   4.7 GB: 8-of-33   4.7 GB: 9-of-33   4.7 GB: 10-of-33   4.7 GB: 11-of-33   4.7 GB: 12-of-33   4.7 GB: 13-of-33   4.7 GB: 14-of-33   4.7 GB: 15-of-33   4.7 GB: 16-of-33   4.7 GB: 17-of-33   4.7 GB: 18-of-33   4.7 GB: 19-of-33   4.7 GB: 20-of-33   4.7 GB: 21-of-33   4.7 GB: 22-of-33   4.7 GB: 23-of-33   4.7 GB: 24-of-33   4.7 GB: 25-of-33   4.7 GB: 26-of-33   4.7 GB: 27-of-33   4.7 GB: 28-of-33   4.7 GB: 29-of-33   4.7 GB: 30-of-33   4.7 GB: 31-of-33   4.7 GB: 32-of-33   1.5 GB: 33-of-33
Supported Languagesen
Quantization Type4bit
Model ArchitectureDeepseekV4ForCausalLM
Context Length1048576
Model Max Length1048576
Transformers Version4.57.1
Tokenizer ClassTokenizersBackend
Padding Token<｜end▁of▁sentence｜>
Vocabulary Size129280
Torch Data Typebfloat16

Best Alternatives to DeepSeek V4 Flash 4bit

Best Alternatives
Context / RAM
Downloads
Likes
DeepSeek V4 Flash 2bit DQ1024K / 96.3 GB4456144
...seek Ai DeepSeek V4 Flash 8bit1024K / 215.2 GB492212
...seek Ai DeepSeek V4 Flash 4bit1024K / 159.7 GB41784
DeepSeek V4 Flash 6bit1024K / 153.5 GB14602
DeepSeek V4 Flash Mxfp81024K / 154.8 GB45413
DeepSeek V4 Flash BF161024K / 568.7 GB3608

Rank the DeepSeek V4 Flash 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? 54415 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