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

Yayi 7B by wenge-research

 »  All LLMs  »  wenge-research  »  Yayi 7B   URL Share it on

  965   

Yayi 7B is an open-source language model by wenge-research. Features: 7b LLM, VRAM: 28.2GB, LLM Explorer Score: 0.12, Arc: 46.3, HellaSwag: 61.7, MMLU: 36.3, GSM8K: 0.9.

  Bloom   En   Endpoints compatible   Pytorch   Region:us   Sharded   Yayi   Zh
Model Card on HF 🤗: https://huggingface.co/wenge-research/yayi-7b 

Yayi 7B Benchmarks

ARC: 46.33 vs 96.7 (so35)-52.1%
HellaSwag: 61.72 vs 95.3 (gpt4)-35.2%
MMLU: 36.34 vs 88.3 (so35)-58.8%
GSM8K: 0.91 vs 96.4 (so35)-99.1%
LLME Score: 0.12024
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Yayi 7B (wenge-research/yayi-7b)
🌟 Advertise your project 🚀

Yayi 7B Parameters and Internals

Model Type 
Text Generation
Use Cases 
Areas:
Research
Applications:
Media publicity, Public opinion analysis, Public safety, Financial risk control, Urban governance
Primary Use Cases:
Over a hundred natural language instruction tasks
Limitations:
Factually incorrect responses, Inability to effectively identify harmful instructions, Requires improvement in logical reasoning, code generation, etc.
Considerations:
Ensure responsible and safe usage in accordance with terms.
Additional Notes 
The model is open-sourced to foster collaboration on the development of Chinese pre-trained models.
Supported Languages 
Chinese (Comprehensive)
Training Details 
Data Sources:
Media publicity data, Public opinion analysis data, Public safety data, Financial risk control data, Urban governance data
Data Volume:
Millions of high-quality domain data
Methodology:
Instruction fine-tuning
Hardware Used:
Single GPU such as A100/A800/3090
Model Architecture:
Pre-trained transformer model architecture
Input Output 
Input Format:
Text with instruction prompts
Accepted Modalities:
Text
Output Format:
Generated text responses
Performance Tips:
Ensure eos_token_id is correctly set for generation.
LLM NameYayi 7B
Repository 🤗https://huggingface.co/wenge-research/yayi-7b 
Model Size7b
Required VRAM28.2 GB
Updated2026-05-28
Maintainerwenge-research
Model Typebloom
Model Files  14.1 GB: 1-of-2   14.1 GB: 2-of-2
Supported Languageszh en
Model ArchitectureBloomForCausalLM
Transformers Version4.28.1
Tokenizer ClassBloomTokenizer
Padding Token<pad>
Vocabulary Size250684
Torch Data Typebfloat16

Best Alternatives to Yayi 7B

Best Alternatives
Context / RAM
Downloads
Likes
Gogpt 7B Bloom0K / 32.3 GB8253
...tral 7B Instruct V0.2 Uld Loss0K / 2.2 GB140
... 7B Instruct V0.2 Text Teacher0K / 2.2 GB130
...Qa Llama 2 7B Chat Hf Uld Loss0K / 2.2 GB60
...lama 2 7B Chat Hf Text Teacher0K / 2.2 GB50
Phoenix Inst Chat 7B0K / 16.2 GB84843
Vietcuna 7B 2k50K / 14.2 GB160
Bloom Xp30K / 32.4 GB50
Tigerbot 7B Sft V20K / 14.2 GB6688
Tigerbot 7B Base V20K / 14.2 GB6373

Rank the Yayi 7B 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? 54358 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