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

Taiwan LLaMa V1.0 4bits GPTQ by q-allen

 »  All LLMs  »  q-allen  »  Taiwan LLaMa V1.0 4bits GPTQ   URL Share it on

  7   
  Arxiv:2104.09864   Arxiv:2305.13711   4-bit   Autotrain compatible Dataset:yentinglin/traditional...   Dataset:yentinglin/zh tw c4   Endpoints compatible   Gptq   Instruct   Llama   Pytorch   Quantized   Region:us   Zh
Model Card on HF 🤗: https://huggingface.co/q-allen/Taiwan-LLaMa-v1.0-4bits-GPTQ 

Taiwan LLaMa V1.0 4bits GPTQ Benchmarks

LLME Score: 0.11308
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Taiwan LLaMa V1.0 4bits GPTQ (q-allen/Taiwan-LLaMa-v1.0-4bits-GPTQ)
🌟 Advertise your project 🚀

Taiwan LLaMa V1.0 4bits GPTQ Parameters and Internals

Model Type 
text generation
Use Cases 
Areas:
Traditional Mandarin applications, Taiwanese culture
Additional Notes 
Pretrained on over 5 billion tokens and instruction-tuned on over 490k conversations in traditional mandarin.
Supported Languages 
zh (Traditional Mandarin)
Training Details 
Data Sources:
zh_TW_c4, traditional_mandarin_instructions
Data Volume:
5 billion tokens
Methodology:
instruction-tuned
Model Architecture:
Transformer architecture, flash attention 2, bfloat16
Input Output 
Input Format:
Prompt format follows vicuna-v1.1 template.
Accepted Modalities:
text
Output Format:
Text generation
Performance Tips:
Advanced customization options like system prompt, temperature, top-p, and top-k.
LLM NameTaiwan LLaMa V1.0 4bits GPTQ
Repository 🤗https://huggingface.co/q-allen/Taiwan-LLaMa-v1.0-4bits-GPTQ 
Model NameLanguage Models for Taiwanese Culture 1.0
Required VRAM7.3 GB
Updated2025-09-23
Maintainerq-allen
Model Typellama
Instruction-BasedYes
Model Files  7.3 GB
Supported Languageszh
GPTQ QuantizationYes
Quantization Typegptq
Model ArchitectureLlamaForCausalLM
Licensellama2
Context Length4096
Model Max Length4096
Transformers Version4.34.1
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to Taiwan LLaMa V1.0 4bits GPTQ

Best Alternatives
Context / RAM
Downloads
Likes
Taiwan LLaMa V1.0 4bits GPTQ4K / 7.3 GB19
Slimorca Phi 3.5128K / 7.6 GB50
Phi 3.5 Instruct Vul128K / 7.6 GB50
...Instruct DPO V2.8.0bpw H8 EXL24K / 10.9 GB52
Reverse Instruct32K / 27 GB63
Codellama Cairo Instruct GGUF16K / 4.1 GB131
Etri Ones Solar4K / 42.9 GB100
Phi 3 Orpo V8.164K / 7.6 GB50
Sakura SOLAR Instruct GGUF4K / 4.5 GB8031
Airavata GGUF4K / 4.2 GB482
Note: green Score (e.g. "73.2") means that the model is better than q-allen/Taiwan-LLaMa-v1.0-4bits-GPTQ.

Rank the Taiwan LLaMa V1.0 4bits GPTQ 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? 51566 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 v20241124
  Support LLM Explorer