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

Nanbeige2 16B Chat by duoqi

 »  All LLMs  »  duoqi  »  Nanbeige2 16B Chat   URL Share it on

  9881   

Nanbeige2 16B Chat is an open-source language model by duoqi. Features: 16b LLM, VRAM: 31.6GB, Context: 4K, License: apache-2.0, HF Score: 67.7, LLM Explorer Score: 0.19, Arc: 63.4, HellaSwag: 80.7, MMLU: 66.6, TruthfulQA: 55, WinoGrande: 76.7, GSM8K: 63.9.

  Autotrain compatible   Conversational   En   Endpoints compatible   Llama   Region:us   Safetensors   Sharded   Tensorflow   Zh
Model Card on HF 🤗: https://huggingface.co/duoqi/Nanbeige2-16B-Chat 

Nanbeige2 16B Chat Benchmarks

ARC: 63.4 vs 96.7 (so35)-34.4%
HellaSwag: 80.7 vs 95.3 (gpt4)-15.3%
MMLU: 66.63 vs 88.3 (so35)-24.5%
TruthfulQA: 55.04 vs 59 (gpt4)-6.7%
WinoGrande: 76.72 vs 87.5 (gpt4)-12.3%
GSM8K: 63.91 vs 96.4 (so35)-33.7%
LLME Score: 0.18996
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Nanbeige2 16B Chat (duoqi/Nanbeige2-16B-Chat)
🌟 Advertise your project 🚀

Nanbeige2 16B Chat Parameters and Internals

Model Type 
text generation
Additional Notes 
The model underwent initial training with 1 million samples through Supervised Fine-Tuning and subsequent curriculum learning on 400,000 challenging samples, with human feedback integrated through Direct Preference Optimization (DPO).
Supported Languages 
en (fluent), zh (fluent)
Training Details 
Data Sources:
4.5T tokens of high-quality training data
Data Volume:
4.5 trillion tokens
Methodology:
Supervised Fine-Tuning (SFT), Curriculum Learning, Direct Preference Optimization (DPO)
LLM NameNanbeige2 16B Chat
Repository 🤗https://huggingface.co/duoqi/Nanbeige2-16B-Chat 
Model Size16b
Required VRAM31.6 GB
Updated2025-07-17
Maintainerduoqi
Model Typellama
Model Files  4.9 GB: 1-of-7   4.9 GB: 2-of-7   4.9 GB: 3-of-7   5.0 GB: 4-of-7   5.0 GB: 5-of-7   4.9 GB: 6-of-7   2.0 GB: 7-of-7
Supported Languagesen zh
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length4096
Model Max Length4096
Transformers Version4.28.1
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size59392
Torch Data Typebfloat16

Best Alternatives to Nanbeige2 16B Chat

Best Alternatives
Context / RAM
Downloads
Likes
...r SMB R1 Distill Llama 3.1 16B128K / 32 GB61
Phi 3.5 Mini Investigator 16B128K / 7.6 GB120
...aid Blackroot Grand HORROR 16B8K / 33.8 GB140
...aid Blackroot Grand HORROR 16B8K / 33.8 GB300
...o Sft Ties Post Merge Auto DPO8K / 141.2 GB570
Nanbeige 16B Base Llama4K / 31.6 GB99464
...ALAXY V03 Slimorca 1 Epoch 50k4K / 31.8 GB1090
...ca 1 Epoch 50k DPO 1 Epoch 30k4K / 31.8 GB790
GALAXY XB V.034K / 31.9 GB60
Llama 2 16B Nastychat4K / 32.4 GB177911
Note: green Score (e.g. "73.2") means that the model is better than duoqi/Nanbeige2-16B-Chat.

Rank the Nanbeige2 16B Chat 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? 52721 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.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer