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

TeleChat 52B by Tele-AI

 »  All LLMs  »  Tele-AI  »  TeleChat 52B   URL Share it on

  29   
  Arxiv:1910.07467   Arxiv:2002.05202   Arxiv:2104.09864   Arxiv:2401.03804   Arxiv:2404.16645   Autotrain compatible   Custom code   Endpoints compatible   Pytorch   Region:us   Sharded   Telechat
Model Card on HF 🤗: https://huggingface.co/Tele-AI/TeleChat-52B 

TeleChat 52B Benchmarks

LLME Score: 0.1877
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
TeleChat 52B (Tele-AI/TeleChat-52B)

TeleChat 52B Parameters and Internals

Model Type 
multilingual, text generation
Use Cases 
Areas:
research, commercial applications
Considerations:
Ensure conformity to safety measures and legal requirements.
Additional Notes 
Model supports multi-round dialog capabilities with optimizations for mask loss training.
Supported Languages 
Chinese (high), English (high)
Training Details 
Data Sources:
Chinese and English high-quality data
Data Volume:
2 trillion tokens
Methodology:
Curriculum learning
Hardware Used:
4x 40G A100 GPUs, DeepSpeed
Model Architecture:
Decoder-only, using Rotary Embedding and SwiGLU activation
Input Output 
Accepted Modalities:
text
Release Notes 
Version:
52B
Date:
5.16
Notes:
Release of the 52B version chat model.
LLM NameTeleChat 52B
Repository 🤗https://huggingface.co/Tele-AI/TeleChat-52B 
Model Size52b
Required VRAM106 GB
Updated2025-02-22
MaintainerTele-AI
Model Typetelechat
Model Files  9.9 GB: 1-of-11   9.7 GB: 2-of-11   9.7 GB: 3-of-11   9.7 GB: 4-of-11   9.7 GB: 5-of-11   9.7 GB: 6-of-11   9.7 GB: 7-of-11   9.7 GB: 8-of-11   9.7 GB: 9-of-11   9.7 GB: 10-of-11   8.8 GB: 11-of-11
Model ArchitectureTeleChatForCausalLM
Model Max Length8192
Transformers Version4.34.1
Tokenizer ClassTELECHATTokenizer
Padding Token<pad>
Vocabulary Size80896
Torch Data Typebfloat16
Activation Functionsilu

Rank the TeleChat 52B 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? 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