LLM EXPLORER 59,420 MODELS INDEXED

GPT J 6B Alpaca Gpt4 by vicgalle

By vicgalle · 46 downloads
HF Repo ↗ Copy URL

GPT J 6B Alpaca Gpt4 is an open-source language model by vicgalle. Features: 6b LLM, VRAM: 24.3GB, License: mit, Quantized, Instruction-Based, LLM Explorer Score: 0.06.

  Alpaca   Dataset:vicgalle/alpaca-gpt4   En   Endpoints compatible   Fp16   Gpt-j   Gpt4   Gptj   Instruct   Pytorch   Quantized   Region:us   Sharded
Model Card on HF 🤗: https://huggingface.co/vicgalle/gpt-j-6B-alpaca-gpt4 

GPT J 6B Alpaca Gpt4 Parameters and Internals

Model Type 
text generation, instruction-tuning
Supported Languages 
en (Proficient)
Training Details 
Data Sources:
vicgalle/alpaca-gpt4
Methodology:
LoRA
Training Time:
over 7 hours
Hardware Used:
four V100S GPUs
Input Output 
Input Format:
prompts with specific instructions
Accepted Modalities:
text
Output Format:
generated text response
LLM NameGPT J 6B Alpaca Gpt4
Repository 🤗https://huggingface.co/vicgalle/gpt-j-6B-alpaca-gpt4 
Model Size6b
Required VRAM24.3 GB
Updated2026-07-13
Maintainervicgalle
Model Typegptj
Instruction-BasedYes
Model Files  10.0 GB: 1-of-3   10.0 GB: 2-of-3   4.3 GB: 3-of-3
Supported Languagesen
Quantization Typefp16
Model ArchitectureGPTJForCausalLM
Licensemit
Model Max Length2048
Transformers Version4.27.4
Tokenizer ClassGPT2Tokenizer
Beginning of Sentence Token<|endoftext|>
End of Sentence Token<|endoftext|>
Unk Token<|endoftext|>
Vocabulary Size50400
Torch Data Typefloat32
Activation Functiongelu_new
Errorsreplace

Best Alternatives to GPT J 6B Alpaca Gpt4

Best Alternatives
Context / RAM
Downloads
Likes
Pyg Instruct Wizardlm0K / 12.2 GB761
Zenos GPT J 6B Instruct 4bit0K / 3.4 GB21
Note: green Score (e.g. "73.2") means that the model is better than vicgalle/gpt-j-6B-alpaca-gpt4.