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

VicUnlocked Alpaca 30B 4bit by Aeala

 »  All LLMs  »  Aeala  »  VicUnlocked Alpaca 30B 4bit   URL Share it on

  7   

VicUnlocked Alpaca 30B 4bit is an open-source language model by Aeala. Features: 30b LLM, VRAM: 16.9GB, Context: 2K, Quantized, HF Score: 57.9, LLM Explorer Score: 0.09, Arc: 61.9, HellaSwag: 83.8, MMLU: 57.6, TruthfulQA: 51, WinoGrande: 78.2, GSM8K: 14.6.

  4bit   Endpoints compatible   Llama   Pytorch   Quantized   Region:us
Model Card on HF 🤗: https://huggingface.co/Aeala/VicUnlocked-alpaca-30b-4bit 

VicUnlocked Alpaca 30B 4bit Benchmarks

ARC: 61.86 vs 96.7 (so35)-36%
HellaSwag: 83.79 vs 95.3 (gpt4)-12.1%
MMLU: 57.64 vs 88.3 (so35)-34.7%
TruthfulQA: 51.03 vs 59 (gpt4)-13.5%
WinoGrande: 78.22 vs 87.5 (gpt4)-10.6%
GSM8K: 14.63 vs 96.4 (so35)-84.8%
LLME Score: 0.09438
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
VicUnlocked Alpaca 30B 4bit (Aeala/VicUnlocked-alpaca-30b-4bit)
🌟 Advertise your project 🚀

VicUnlocked Alpaca 30B 4bit Parameters and Internals

Model Type 
Text Generation
Additional Notes 
Use Alpaca-style prompts for optimal results.
Training Details 
Data Sources:
ShareGPT dataset, Alpaca format
Data Volume:
Unknown
Methodology:
4-bit GPTQ quantization
Model Architecture:
Alpaca format based on Vicuna model
Input Output 
Input Format:
### Instruction: (without the <>) ### Response:
Accepted Modalities:
text
Output Format:
text
LLM NameVicUnlocked Alpaca 30B 4bit
Repository 🤗https://huggingface.co/Aeala/VicUnlocked-alpaca-30b-4bit 
Base Model(s)  VicUnlocked Alpaca 30B   Aeala/VicUnlocked-alpaca-30b
Model Size30b
Required VRAM16.9 GB
Updated2026-03-31
MaintainerAeala
Model Typellama
Model Files  18.1 GB   16.9 GB
Quantization Type4bit
Model ArchitectureLlamaForCausalLM
Context Length2048
Model Max Length2048
Transformers Version4.29.0
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to VicUnlocked Alpaca 30B 4bit

Best Alternatives
Context / RAM
Downloads
Likes
Tenebra 30B Alpha01 FP1616K / 65 GB18339
Tenebra 30B Alpha01 4BIT16K / 19.4 GB1161
...nebra 30B Alpha01 EXL2 2 80bpw16K / 11.9 GB41
...nebra 30B Alpha01 EXL2 2 50bpw16K / 10.7 GB70
Tenebra 30B Alpha01 EXL2 3bpw16K / 12.7 GB60
Tenebra 30B Alpha01 3BIT16K / 12.9 GB60
Platypus 30B SuperHOT 8K Fp168K / 65.2 GB11692
... 30B Supercot SuperHOT 8K Fp168K / 65.2 GB10774
GPlatty 30B SuperHOT 8K Fp168K / 65.2 GB10641
Tulu 30B Fp162K / 65.2 GB19285
Note: green Score (e.g. "73.2") means that the model is better than Aeala/VicUnlocked-alpaca-30b-4bit.

Rank the VicUnlocked Alpaca 30B 4bit 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? 53407 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