Chatglm2 6B by THUDM

 ยป  All LLMs  ยป  THUDM  ยป  Chatglm2 6B   URL Share it on

  Arxiv:1911.02150   Arxiv:2103.10360   Arxiv:2210.02414   Arxiv:2406.12793   Chatglm   Custom code   En   Endpoints compatible   Glm   Pytorch   Region:us   Sharded   Thudm   Zh
Model Card on HF ๐Ÿค—: https://huggingface.co/THUDM/chatglm2-6b 

Chatglm2 6B Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Chatglm2 6B (THUDM/chatglm2-6b)

Chatglm2 6B Parameters and Internals

Model Type 
Bilingual, Chatbot
Use Cases 
Areas:
Research, Commercial Applications
Applications:
Conversational AI, Bilingual Chatbot
Primary Use Cases:
Dialog Generation, Language Understanding
Limitations:
Limited understanding of ultra-long documents
Additional Notes 
Improved context length, inference speed, and model efficiency.
Supported Languages 
Chinese (Bilingual), English (Bilingual)
Training Details 
Data Sources:
1.4T bilingual tokens
Methodology:
Pre-training with a hybrid objective function and human preference alignment training
Context Length:
32000
Model Architecture:
GLM architecture with FlashAttention and Multi-Query Attention
Input Output 
Accepted Modalities:
text
Release Notes 
Version:
v1.0
Notes:
Initial release with several key improvements over the first-generation model.
LLM NameChatglm2 6B
Repository ๐Ÿค—https://huggingface.co/THUDM/chatglm2-6b 
Model Size6b
Required VRAM12.5 GB
Updated2025-02-05
MaintainerTHUDM
Model Typechatglm
Model Files  1.8 GB: 1-of-7   2.0 GB: 2-of-7   1.9 GB: 3-of-7   1.8 GB: 4-of-7   2.0 GB: 5-of-7   1.9 GB: 6-of-7   1.1 GB: 7-of-7
Supported Languageszh en
Model ArchitectureChatGLMModel
Transformers Version4.27.1
Tokenizer ClassChatGLMTokenizer
Torch Data Typefloat16

Best Alternatives to Chatglm2 6B

Best Alternatives
Context / RAM
Downloads
Likes
Chatglm3 6B0K / 12.5 GB304621105
Chatglm 6B0K / 13.5 GB86632850
Codegeex2 6B0K / 12.5 GB168253
Octogeex0K / 12.5 GB10822

Rank the Chatglm2 6B 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? 42577 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227