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

YuLan Chat 3 12B by yulan-team

 »  All LLMs  »  yulan-team  »  YuLan Chat 3 12B   URL Share it on

  11   
  Arxiv:2406.19853   Autotrain compatible   Endpoints compatible   Llama   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/yulan-team/YuLan-Chat-3-12b 

YuLan Chat 3 12B Benchmarks

LLME Score: 0.16303
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
YuLan Chat 3 12B (yulan-team/YuLan-Chat-3-12b)
🌟 Advertise your project 🚀

YuLan Chat 3 12B Parameters and Internals

Model Type 
chat-based, large language model
Use Cases 
Areas:
academic, bilingual chatbot
Applications:
chatbots, language understanding
Primary Use Cases:
chat applications, bilingual assistance
Limitations:
may produce unexpected outputs, probabilistic generation can lead to biases or discrimination
Considerations:
Use with caution for generating ethical and aligned content.
Additional Notes 
Bilingual focus improves model adaptability to both English and Chinese cultures.
Supported Languages 
Chinese (high proficiency), English (high proficiency), Multilingual (moderate proficiency)
Training Details 
Data Sources:
over 1.6TB tokens of English, Chinese, multilingual data
Data Volume:
1.6TB
Methodology:
Supervised fine-tuning via curriculum learning
Context Length:
4096
Model Architecture:
Based on LLaMA and LLaMA-2
Safety Evaluation 
Risk Categories:
bias, discrimination
Ethical Considerations:
Please do not propagate harmful content generated by the model.
Responsible Ai Considerations 
Fairness:
Efforts made to reduce potential biases and discrimination.
Transparency:
Model weights and differences provided.
Accountability:
Users are responsible for avoiding dissemination of harmful content.
Mitigation Strategies:
Encouragement to generate ethical and legal text.
Input Output 
Input Format:
Supports up to 4096 context tokens.
Accepted Modalities:
text
Output Format:
text
Performance Tips:
Considerations for handling large input/output safely.
Release Notes 
Version:
YuLan-Base-12B
Date:
2024.7.1
Notes:
Base model trained from scratch with bilingual data.
Version:
YuLan-Chat-3-12B
Date:
2024.7.1
Notes:
Chat-based version through fine-tuning.
Version:
YuLan-Chat-2-13B
Date:
2023.8.2
Notes:
Improved language abilities pre-trained on LLAMA-2.
Version:
YuLan-Chat-1-65B-v2
Date:
2023.8.2
Notes:
Includes advancements in vocabulary and processing.
Version:
YuLan-Chat-1-13B-v1
Date:
2023.6.8
Notes:
Initial release of the chat model series.
LLM NameYuLan Chat 3 12B
Repository 🤗https://huggingface.co/yulan-team/YuLan-Chat-3-12b 
Model Size12b
Required VRAM23.8 GB
Updated2025-09-27
Maintaineryulan-team
Model Typellama
Model Files  4.9 GB: 1-of-5   4.9 GB: 2-of-5   5.0 GB: 3-of-5   5.0 GB: 4-of-5   4.0 GB: 5-of-5   0.0 GB
Model ArchitectureLlamaForCausalLM
Licensemit
Context Length4096
Model Max Length4096
Transformers Version4.41.1
Tokenizer ClassLlamaTokenizer
Vocabulary Size51190
Torch Data Typebfloat16

Best Alternatives to YuLan Chat 3 12B

Best Alternatives
Context / RAM
Downloads
Likes
L3.1 Smth.Concv3 12B128K / 47.8 GB560
L3.1 Legionv2 12B128K / 47.8 GB200
L3.1 Nimbusv1 12B128K / 48.1 GB70
L3.1 Adumbralv1 12B128K / 47.8 GB60
L3.1 Punicav1 12B128K / 48.1 GB120
L3.1 Twilightv1 12B128K / 48.1 GB60
OpenCrystal 12B L3.1 128K128K / 23 GB43
OpenCrystal 12B L38K / 23 GB1717
Llama3 12B8K / 23.1 GB121
IxChel L3 12B8K / 23 GB42
Note: green Score (e.g. "73.2") means that the model is better than yulan-team/YuLan-Chat-3-12b.

Rank the YuLan Chat 3 12B 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? 51544 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