Baichuan 13B Instruction GPTQ by s3nh

 ยป  All LLMs  ยป  s3nh  ยป  Baichuan 13B Instruction GPTQ   URL Share it on

  Arxiv:2104.09864   Arxiv:2108.12409   Arxiv:2307.00360   4bit   Autotrain compatible   Baichuan   Custom code   En   Gptq   Instruct   Quantized   Region:us

Baichuan 13B Instruction GPTQ Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Baichuan 13B Instruction GPTQ (s3nh/Baichuan-13B-Instruction-GPTQ)

Baichuan 13B Instruction GPTQ Parameters and Internals

Model Type 
text-generation
Additional Notes 
The model supports int8 and int4 quantization for reduced inference memory footprint.
Supported Languages 
Chinese (Advanced), English (Intermediate)
Training Details 
Data Sources:
sharegpt_zh, lima, custom Chinese dataset
Methodology:
ALiBi linear bias technique
Context Length:
4096
Hardware Used:
8*A40 GPUs
Model Architecture:
Based on Baichuan-13B with optimization for inference using ALiBi and compared with LLaMA-13B
Input Output 
Input Format:
Dictionary with role and content keys
Accepted Modalities:
text
Output Format:
Text response
Performance Tips:
Load the model on CPU before quantizing to int8 or int4 to save memory.
LLM NameBaichuan 13B Instruction GPTQ
Repository ๐Ÿค—https://huggingface.co/s3nh/Baichuan-13B-Instruction-GPTQ 
Base Model(s)  Baichuan 13B Instruction   AlpachinoNLP/Baichuan-13B-Instruction
Model Size13b
Required VRAM7.9 GB
Updated2025-04-29
Maintainers3nh
Model Typebaichuan
Instruction-BasedYes
Model Files  7.9 GB
Supported Languagesen
GPTQ QuantizationYes
Quantization Typegptq|4bit
Model ArchitectureBaichuanForCausalLM
Licenseopenrail
Model Max Length4096
Transformers Version4.31.0
Vocabulary Size64000
Torch Data Typefloat16

Best Alternatives to Baichuan 13B Instruction GPTQ

Best Alternatives
Context / RAM
Downloads
Likes
... Efficient Training Of LLMs V10K / 29.1 GB111
NSFW 13B Sft0K / 26.5 GB8295
Blossom V2 Baichuan 13B0K / 26.5 GB101
Baichuan 13B Instruction0K / 26.5 GB106
Note: green Score (e.g. "73.2") means that the model is better than s3nh/Baichuan-13B-Instruction-GPTQ.

Rank the Baichuan 13B Instruction GPTQ 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? 46792 in total.

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