What are the hardware requirements for KoAlpaca 30B LoRA?

KoAlpaca 30B LoRA requires approximately 0.1 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed KoAlpaca 30B LoRA and how large is it?

KoAlpaca 30B LoRA is developed by beomi, a model with 30b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate KoAlpaca 30B LoRA?

KoAlpaca 30B LoRA is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

KoAlpaca 30B LoRA by beomi — VRAM 0.1GB

Name: KoAlpaca 30B LoRA
Author: beomi

KoAlpaca 30B LoRA is an open-source language model by beomi. Features: 30b LLM, VRAM: 0.1GB, License: mit, LLM Explorer Score: 0.06.

Adapter Alpaca En Finetuned Ko Koalpaca Llama Lora Region:us

Model Card on HF 🤗: https://huggingface.co/beomi/KoAlpaca-30B-LoRA

KoAlpaca 30B LoRA Parameters and Internals

Model Type

finetuned, LoRA

Additional Notes

The repository contains only the LoRA weights, not the full model weights.

Supported Languages

ko (high), en (high)

Training Details

Data Sources:

Korean dataset, English dataset

Methodology:

Finetuning (LoRA)

LLM Name	KoAlpaca 30B LoRA
Repository 🤗	https://huggingface.co/beomi/KoAlpaca-30B-LoRA
Model Size	30b
Required VRAM	0.1 GB
Updated	2026-08-02
Maintainer	beomi
Model Files	0.1 GB
Supported Languages	ko en
Model Architecture	Adapter
License	mit
Is Biased	none
PEFT Type	LORA
LoRA Model	Yes
PEFT Target Modules	q_proj\|v_proj
LoRA Alpha	16
LoRA Dropout	0.05
R Param	8

Rank the KoAlpaca 30B LoRA capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

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

Best Alternatives to KoAlpaca 30B LoRA

Best Alternatives	Context / RAM	Downloads	Likes
...3.5 30B A3B Antislop FTPO LoRA	0K / 0.8 GB	7	1
...ans PersonalityEngine 30B LoRA	0K / 0.2 GB	0	1
...ans PersonalityEngine 30B LoRA	0K / 0.2 GB	0	1
SoTana 30B Lora 500000	0K / 0.1 GB	0	1
...cUnlocked Alpaca Half 30B LoRA	0K / 0.2 GB	0	9
Gpt4 Alpaca Lora 30B	0K / 0.2 GB	0	64
Saiga 30b Lora	0K / 0.2 GB	0	23
Tianpeng Lora 30B	0K / 0.2 GB	0	7
Baize Lora 30B	0K / 0.2 GB	0	34
Alpaca Lora 30B	0K / 0.2 GB	0	49

Note: green Score (e.g. "73.2") means that the model is better than beomi/KoAlpaca-30B-LoRA.