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

Chinese Llama 2 7B 4bit by LinkSoul

 »  All LLMs  »  LinkSoul  »  Chinese Llama 2 7B 4bit   URL Share it on

  19   

Chinese Llama 2 7B 4bit is an open-source language model by LinkSoul. Features: 7b LLM, VRAM: 13.5GB, Context: 2K, License: openrail, Quantized, Instruction-Based, Merged, LLM Explorer Score: 0.12, Arc: 44.5, HellaSwag: 69.5, MMLU: 37.4, GSM8K: 3.8.

  Merged Model   4bit Dataset:linksoul/instruction m...   En   Endpoints compatible   Instruct   Llama   Pytorch   Quantized   Region:us   Sharded   Zh
Model Card on HF 🤗: https://huggingface.co/LinkSoul/Chinese-Llama-2-7b-4bit 

Chinese Llama 2 7B 4bit Benchmarks

ARC: 44.45 vs 96.7 (so35)-54%
HellaSwag: 69.54 vs 95.3 (gpt4)-27%
MMLU: 37.42 vs 88.3 (so35)-57.6%
GSM8K: 3.79 vs 96.4 (so35)-96.1%
LLME Score: 0.12133
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Chinese Llama 2 7B 4bit (LinkSoul/Chinese-Llama-2-7b-4bit)
🌟 Advertise your project 🚀

Chinese Llama 2 7B 4bit Parameters and Internals

Model Type 
text generation, multimodal
Use Cases 
Areas:
research, commercial
Applications:
text generation
Primary Use Cases:
multilingual chatbots in Chinese and English
Additional Notes 
Compatible with all optimizations for the original Llama-2-chat model
Supported Languages 
zh (high), en (moderate)
Training Details 
Data Sources:
LinkSoul/instruction_merge_set
Data Volume:
10 million
Model Architecture:
Llama-2-chat with optimizations
Input Output 
Input Format:
Llama-2-chat format with instructions
Accepted Modalities:
text
Output Format:
text
LLM NameChinese Llama 2 7B 4bit
Repository 🤗https://huggingface.co/LinkSoul/Chinese-Llama-2-7b-4bit 
Base Model(s)  Chinese Llama 2 7B 64K   hfl/chinese-llama-2-7b-64k
Merged ModelYes
Model Size7b
Required VRAM13.5 GB
Updated2026-05-20
MaintainerLinkSoul
Model Typellama
Instruction-BasedYes
Model Files  10.0 GB: 1-of-2   3.5 GB: 2-of-2
Supported Languageszh en
Quantization Type4bit
Model ArchitectureLlamaForCausalLM
Licenseopenrail
Context Length2048
Model Max Length2048
Transformers Version4.30.2
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to Chinese Llama 2 7B 4bit

Best Alternatives
Context / RAM
Downloads
Likes
...p 0.05 Max Grad1.0 Grad Accu3232K / 14.4 GB90
...p 0.05 Max Grad1.0 Grad Accu3232K / 14.4 GB60
...ruct Solidity Bnb 4bit Smashed16K / 4.2 GB60
...B Instruct Hf Bnb 4bit Smashed16K / 4.2 GB60
...eLlama 7B Instruct Hf 4bit MLX16K / 4.2 GB1053
CodelLama7B Inst DPO 7K Mlx16K / 4.2 GB234
...6.7B Instruct 8.0bpw H8 EXL2 216K / 6.8 GB72
...6.7B Instruct 3.0bpw H6 EXL2 216K / 2.8 GB41
...coder S CL 7B 3.0bpw H6 EXL2 216K / 2.8 GB71
...coder S CL 7B 5.0bpw H6 EXL2 216K / 4.4 GB31
Note: green Score (e.g. "73.2") means that the model is better than LinkSoul/Chinese-Llama-2-7b-4bit.

Rank the Chinese Llama 2 7B 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? 53972 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