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

LongAlign 6B 64K Base by THUDM

 ยป  All LLMs  ยป  THUDM  ยป  LongAlign 6B 64K Base   URL Share it on

  28   
  Arxiv:2401.18058   Chatglm   Custom code   En   Feature-extraction   Long context   Pytorch   Region:us   Sharded   Zh
Model Card on HF ๐Ÿค—: https://huggingface.co/THUDM/LongAlign-6B-64k-base 

LongAlign 6B 64K Base Benchmarks

LLME Score: 0.16201
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
LongAlign 6B 64K Base (THUDM/LongAlign-6B-64k-base)

LongAlign 6B 64K Base Parameters and Internals

Model Type 
text-generation
Use Cases 
Areas:
text generation
Applications:
chatbot applications
Primary Use Cases:
long context alignment, instruction-following capability
Additional Notes 
LongAlign offers a unique alignment recipe for LLMs with emphasis on long context understanding.
Supported Languages 
en (full), zh (full)
Training Details 
Data Sources:
LongAlign-10k dataset
Data Volume:
10,000 long instruction data of 8k-64k in length
Methodology:
packing (with loss weighting) and sorted batching
Context Length:
64000
Input Output 
Input Format:
[Round n] ้—ฎ๏ผš{question} ็ญ”๏ผš{answer}
Accepted Modalities:
text
Output Format:
Generated text response
LLM NameLongAlign 6B 64K Base
Repository ๐Ÿค—https://huggingface.co/THUDM/LongAlign-6B-64k-base 
Model Size6b
Required VRAM12.3 GB
Updated2025-02-22
MaintainerTHUDM
Model Typechatglm
Model Files  0.4 GB: 0-of-29   0.4 GB: 1-of-29   0.4 GB: 10-of-29   0.4 GB: 11-of-29   0.4 GB: 12-of-29   0.4 GB: 13-of-29   0.4 GB: 14-of-29   0.4 GB: 15-of-29   0.4 GB: 16-of-29   0.4 GB: 17-of-29   0.4 GB: 18-of-29   0.4 GB: 19-of-29   0.4 GB: 2-of-29   0.4 GB: 20-of-29   0.4 GB: 21-of-29   0.4 GB: 22-of-29   0.4 GB: 23-of-29   0.4 GB: 24-of-29   0.4 GB: 25-of-29   0.4 GB: 26-of-29   0.4 GB: 27-of-29   1.1 GB: 28-of-29   0.4 GB: 3-of-29   0.4 GB: 4-of-29   0.4 GB: 5-of-29   0.4 GB: 6-of-29   0.4 GB: 7-of-29   0.4 GB: 8-of-29   0.4 GB: 9-of-29
Supported Languagesen zh
Model ArchitectureChatGLMForConditionalGeneration
Licenseapache-2.0
Transformers Version4.33.0
Tokenizer ClassChatGLMTokenizer
Vocabulary Size65024
Torch Data Typebfloat16

Best Alternatives to LongAlign 6B 64K Base

Best Alternatives
Context / RAM
Downloads
Likes
ChatGLM3 6B Theresa0K / 12.5 GB91
LongAlign 6B 64K0K / 12.5 GB422
Note: green Score (e.g. "73.2") means that the model is better than THUDM/LongAlign-6B-64k-base.

Rank the LongAlign 6B 64K Base Capabilities

๐Ÿ†˜ 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! ๐ŸŒŸ

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? 43470 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 v20241227
  Support LLM Explorer