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

Llama3 8B Chinese Chat GGUF 4bit by shenzhi-wang

 »  All LLMs  »  shenzhi-wang  »  Llama3 8B Chinese Chat GGUF 4bit   URL Share it on

  852   
  Arxiv:2403.07691   4bit   Autotrain compatible Base model:meta-llama/meta-lla... Base model:quantized:meta-llam...   Conversational   En   Endpoints compatible   Gguf   Instruct   Llama   Llama-factory   Orpo   Q4   Q4 0   Quantized   Region:us   Zh
Model Card on HF 🤗: https://huggingface.co/shenzhi-wang/Llama3-8B-Chinese-Chat-GGUF-4bit 

Llama3 8B Chinese Chat GGUF 4bit Benchmarks

LLME Score: 0.22224
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Llama3 8B Chinese Chat GGUF 4bit (shenzhi-wang/Llama3-8B-Chinese-Chat-GGUF-4bit)

Llama3 8B Chinese Chat GGUF 4bit Parameters and Internals

Model Type 
text-generation
Use Cases 
Primary Use Cases:
roleplaying, tool-using, math, Chinese-English user assistance
Additional Notes 
Compared to the original Meta-Llama-3-8B-Instruct model, the Llama3-8B-Chinese-Chat model significantly reduces issues of "Chinese questions with English answers."
Supported Languages 
en (proficient), zh (proficient)
Training Details 
Methodology:
{ "[1] Hong, Jiwoo, Noah Lee, and James Thorne. "Reference-free Monolithic Preference Optimization with Odds Ratio." arXiv preprint arXiv:2403.07691 (2024).
Context Length:
8192
Model Architecture:
Based on Meta-Llama-3-8B-Instruct. The model is tuned to reduce the issues of "Chinese questions with English answers" and the mixing of Chinese and English in responses.
Release Notes 
Version:
v2.1
Date:
May 6, 2024
Notes:
- Compared to v1, the training dataset of v2.1 is 5x larger (~100K preference pairs). - Significant enhancements in roleplay, function calling, and math capabilities. - Surpasses v2 in math and is less prone to including English words in Chinese responses.
Version:
v2
Date:
April 29, 2024
Notes:
- Compared to v1, the training dataset of v2 is 5x larger (~100K preference pairs). - Enhancements in roleplay, tool using, and math capabilities.
LLM NameLlama3 8B Chinese Chat GGUF 4bit
Repository 🤗https://huggingface.co/shenzhi-wang/Llama3-8B-Chinese-Chat-GGUF-4bit 
Base Model(s)  Meta Llama 3 8B Instruct   meta-llama/Meta-Llama-3-8B-Instruct
Model Size8b
Required VRAM4.7 GB
Updated2024-12-26
Maintainershenzhi-wang
Model Typellama
Instruction-BasedYes
Model Files  4.7 GB
Supported Languagesen zh
GGUF QuantizationYes
Quantization Typegguf|q4|q4_0|4bit
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length8192
Model Max Length8192
Transformers Version4.40.0
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Llama3 8B Chinese Chat GGUF 4bit

Best Alternatives
Context / RAM
Downloads
Likes
...truct Gradient 1048K IMat GGUF1024K / 2 GB2726
...B Instruct Gradient 1048K GGUF1024K / 3.2 GB1623
Llama 3 8B Instruct 262K GGUF256K / 3.2 GB1582
... 8B Instruct Reasoner 1o1 V0.3128K / 16.1 GB2437
...lama 3.1 Cantonese 8B Instruct128K / 16.1 GB5165
ProductLlama V2128K / 16.1 GB260
Alpha R S V2 Q8 0 GGUF39K / 8.5 GB90
SmolTulu 1.7B Instruct8K / 3.4 GB34813
Llama 3 Cantonese 8B Instruct8K / 16.1 GB6815
...ama3 8B Chinese Chat GGUF 8bit8K / 8.5 GB709166
Note: green Score (e.g. "73.2") means that the model is better than shenzhi-wang/Llama3-8B-Chinese-Chat-GGUF-4bit.

Rank the Llama3 8B Chinese Chat GGUF 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? 40248 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 v20241217
  Support LLM Explorer