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

DeepSeek V2 Lite by deepseek-ai

 »  All LLMs  »  deepseek-ai  »  DeepSeek V2 Lite   URL Share it on

  76774   
  Arxiv:2405.04434   Autotrain compatible   Conversational   Custom code   Deepseek v2   Endpoints compatible   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/deepseek-ai/DeepSeek-V2-Lite 

DeepSeek V2 Lite Benchmarks

LLME Score: 0.23726
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
DeepSeek V2 Lite (deepseek-ai/DeepSeek-V2-Lite)
🌟 Advertise your project 🚀

DeepSeek V2 Lite Parameters and Internals

Model Type 
Mixture-of-Experts, Language Model
Additional Notes 
MLA guarantees efficient inference through significantly compressing the Key-Value (KV) cache into a latent vector.
Supported Languages 
English (High), Chinese (High)
Training Details 
Data Volume:
5.7T tokens
Methodology:
Multi-head Latent Attention, DeepSeekMoE
Context Length:
128000
Model Architecture:
27 layers, hidden dimension of 2048, MLA, 16 attention heads, KV compression to latent vector.
Input Output 
Accepted Modalities:
text
Release Notes 
Version:
DeepSeek-V2
Date:
2024-05-06
Notes:
Released the main model with 236B total parameters.
Version:
DeepSeek-V2-Lite
Date:
2024-05-16
Notes:
Introduced a smaller version with 16B total params.
LLM NameDeepSeek V2 Lite
Repository 🤗https://huggingface.co/deepseek-ai/DeepSeek-V2-Lite 
Model Size15.7b
Required VRAM31.4 GB
Updated2025-09-23
Maintainerdeepseek-ai
Model Typedeepseek_v2
Model Files  8.6 GB: 1-of-4   8.6 GB: 2-of-4   8.6 GB: 3-of-4   5.6 GB: 4-of-4
Model ArchitectureDeepseekV2ForCausalLM
Licenseother
Context Length163840
Model Max Length163840
Transformers Version4.33.1
Tokenizer ClassLlamaTokenizerFast
Beginning of Sentence Token<｜begin▁of▁sentence｜>
End of Sentence Token<｜end▁of▁sentence｜>
Vocabulary Size102400
Torch Data Typebfloat16

Best Alternatives to DeepSeek V2 Lite

Best Alternatives
Context / RAM
Downloads
Likes
...eepSeek Coder V2 Lite Instruct160K / 31.4 GB541650473
...e Sft Math7k 1epoch 1e 4 Gamma160K / 31.4 GB3270
DeepSeek V2 Lite Chat160K / 31.4 GB85172131
Dsv2 Lite Coder Codeio Pp160K / 31.4 GB112
DeepSeek Coder V2 Lite Base160K / 31.4 GB296090
KwaiCoder DS V2 Lite Base160K / 31.4 GB74
...2 Lite Instruct FlashAttnPatch160K / 31.4 GB50
ESFT Vanilla Lite4K / 31.4 GB110818
LunarAI160K / 31.4 GB130
Note: green Score (e.g. "73.2") means that the model is better than deepseek-ai/DeepSeek-V2-Lite.

Rank the DeepSeek V2 Lite 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? 51545 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.
Release v20241124
  Support LLM Explorer