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

Ziya LLaMA 13B V1 by IDEA-CCNL

 »  All LLMs  »  IDEA-CCNL  »  Ziya LLaMA 13B V1   URL Share it on

  1814   
  Arxiv:2210.08590   Autotrain compatible   En   Llama   Pytorch   Region:us   Sharded   Zh
Model Card on HF 🤗: https://huggingface.co/IDEA-CCNL/Ziya-LLaMA-13B-v1 

Ziya LLaMA 13B V1 Benchmarks

MMLU Pro: 1.12
MUSR: 3.88
BBH: 1.46
IFEval: 16.97 vs 88 (so35)-80.7%
ARC: 62.4 vs 96.7 (so35)-35.5%
HellaSwag: 76.9 vs 95.3 (gpt4)-19.3%
MMLU: 50.3 vs 88.3 (so35)-43%
TruthfulQA: 48.65 vs 59 (gpt4)-17.5%
WinoGrande: 49.57 vs 87.5 (gpt4)-43.3%
GSM8K: 40.4 vs 96.4 (so35)-58.1%
OpenCompas: 48.5 vs 73.3 (gpt4)-33.8%
Exam: 45.8 vs 77.2 (gpt4)-40.7%
Language: 37.5 vs 62 (gpt4)-39.5%
Knowledge: 58.4 vs 73.5 (gpt4)-20.5%
Understand: 59.6 vs 70 (gpt4)-14.9%
Reasoning: 45.7 vs 74.4 (gpt4)-38.6%
LLME Score: 0.17202
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Ziya LLaMA 13B V1 (IDEA-CCNL/Ziya-LLaMA-13B-v1)
🌟 Advertise your project 🚀

Ziya LLaMA 13B V1 Parameters and Internals

Model Type 
large-scale pre-trained, text generation
Use Cases 
Areas:
Research, Non-commercial applications
Applications:
Translation, Programming, Text classification, Information extraction, Summarization, Copywriting, Common sense Q&A, Mathematics
Limitations:
Cannot be used for commercial purposes due to LLaMA license
Additional Notes 
Model weights differences published due to licensing restrictions, requiring user intervention to obtain complete weights.
Supported Languages 
en (fluent), zh (fluent)
Training Details 
Data Sources:
openwebtext, Books, Wikipedia, Code, Cleaned Wudao dataset, self-built Chinese dataset
Data Volume:
125 billion tokens
Methodology:
Continual pretraining, multi-task supervised fine-tuning, human feedback learning
Training Time:
8 days
Hardware Used:
160 A100 GPUs, 40GB each
Model Architecture:
Enhanced LLaMA with 8,000 Chinese characters
Responsible Ai Considerations 
Transparency:
Loss curve during training released to help understand potential issues.
Input Output 
Input Format:
Tokenized text (LlamaTokenizer)
Accepted Modalities:
text
Output Format:
Generated text
LLM NameZiya LLaMA 13B V1
Repository 🤗https://huggingface.co/IDEA-CCNL/Ziya-LLaMA-13B-v1 
Model Size13b
Required VRAM26.1 GB
Updated2025-09-15
MaintainerIDEA-CCNL
Model Typellama
Model Files  0.9 GB: 1-of-28   1.0 GB: 2-of-28   0.9 GB: 3-of-28   1.0 GB: 4-of-28   0.9 GB: 5-of-28   1.0 GB: 6-of-28   0.9 GB: 7-of-28   1.0 GB: 8-of-28   0.9 GB: 9-of-28   1.0 GB: 10-of-28   0.9 GB: 11-of-28   1.0 GB: 12-of-28   0.9 GB: 13-of-28   1.0 GB: 14-of-28   0.9 GB: 15-of-28   1.0 GB: 16-of-28   0.9 GB: 17-of-28   1.0 GB: 18-of-28   0.9 GB: 19-of-28   1.0 GB: 20-of-28   0.9 GB: 21-of-28   1.0 GB: 22-of-28   0.9 GB: 23-of-28   1.0 GB: 24-of-28   0.9 GB: 25-of-28   1.0 GB: 26-of-28   0.9 GB: 27-of-28   0.5 GB: 28-of-28
Supported Languagesen zh
Model ArchitectureLlamaForCausalLM
Licensegpl-3.0
Context Length2048
Model Max Length2048
Transformers Version4.29.0.dev0
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size39424
Torch Data Typefloat16

Best Alternatives to Ziya LLaMA 13B V1

Best Alternatives
Context / RAM
Downloads
Likes
Luminaura RP 13B128K / 26 GB50
Yarn Llama 2 13B 128K128K / 26 GB200112
Agent Llama2 13B 80K80K / 26.4 GB80
Chat Llama2 13B 80K80K / 52.8 GB80
LongAlign 13B 64K64K / 26 GB11613
LongAlign 13B 64K Base64K / 26 GB923
LongAlign 13B 64K64K / 26 GB1113
LongAlign 13B 64K Base64K / 26 GB63
Openbuddy Llama2 13B V15p1 64K64K / 26.1 GB34
Openbuddy Llama2 13b64k V1564K / 26.1 GB32

Rank the Ziya LLaMA 13B V1 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? 51387 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