LLM EXPLORER 59,278 MODELS INDEXED

TeleChat3 36B Thinking by Tele-AI

By Tele-AI · 18267 downloads
HF Repo ↗ Copy URL

TeleChat3 36B Thinking is an open-source language model by Tele-AI. Features: 36b LLM, VRAM: 72.8GB, Context: 32K, LLM Explorer Score: 0.28.

  Arxiv:2507.18013   Custom code   Region:us   Safetensors   Sharded   Telechat3   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Tele-AI/TeleChat3-36B-Thinking 

TeleChat3 36B Thinking Parameters and Internals

LLM NameTeleChat3 36B Thinking
Repository 🤗https://huggingface.co/Tele-AI/TeleChat3-36B-Thinking 
Model Size36b
Required VRAM72.8 GB
Updated2026-08-02
MaintainerTele-AI
Model Typetelechat3
Model Files  4.0 GB: 1-of-20   3.9 GB: 2-of-20   3.7 GB: 3-of-20   3.9 GB: 4-of-20   3.7 GB: 5-of-20   3.9 GB: 6-of-20   3.7 GB: 7-of-20   3.9 GB: 8-of-20   3.7 GB: 9-of-20   3.9 GB: 10-of-20   3.7 GB: 11-of-20   3.9 GB: 12-of-20   3.7 GB: 13-of-20   3.9 GB: 14-of-20   3.7 GB: 15-of-20   3.9 GB: 16-of-20   3.7 GB: 17-of-20   3.9 GB: 18-of-20   2.5 GB: 19-of-20   1.6 GB: 20-of-20
Model ArchitectureTeleChat3ForCausalLM
Context Length32768
Model Max Length32768
Transformers Version4.53.2
Tokenizer ClassTelechat3Tokenizer
Padding Token<_pad>
Vocabulary Size131072
Torch Data Typebfloat16