What are the hardware requirements for KoLLaVA KoVicuna 7B?

KoLLaVA KoVicuna 7B requires approximately 27 GB of VRAM and supports a context window of 2K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed KoLLaVA KoVicuna 7B and how large is it?

KoLLaVA KoVicuna 7B is developed by tabtoyou, a model with 7b 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 KoLLaVA KoVicuna 7B?

KoLLaVA KoVicuna 7B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

KoLLaVA KoVicuna 7B by tabtoyou — VRAM 27GB, 2K context

Name: KoLLaVA KoVicuna 7B
Author: tabtoyou

KoLLaVA KoVicuna 7B is an open-source language model by tabtoyou. Features: 7b LLM, VRAM: 27GB, Context: 2K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.07.

Clip Dataset:tabtoyou/kollava-cc3m-... Dataset:tabtoyou/kollava-instr... Endpoints compatible Instruct Ko Koalpaca Kollava Kovicuna Llava Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/tabtoyou/KoLLaVA-KoVicuna-7b

KoLLaVA KoVicuna 7B Parameters and Internals

Model Type

multimodal

Additional Notes

The model combines KoVicuna with a visual encoder from CLIP (ViT-14) and is trained on Korean visual-instruction datasets.

Supported Languages

ko (Korean)

Training Details

Data Sources:

tabtoyou/KoLLaVA-Instruct-150k, tabtoyou/KoLLaVA-CC3M-Pretrain-595K

Hardware Used:

Multi-GPU (A100 80G)

Input Output

Accepted Modalities:

text, image

LLM Name	KoLLaVA KoVicuna 7B
Repository 🤗	https://huggingface.co/tabtoyou/KoLLaVA-KoVicuna-7b
Model Size	7b
Required VRAM	27 GB
Updated	2026-06-15
Maintainer	tabtoyou
Model Type	llava
Instruction-Based	Yes
Model Files	9.9 GB: 1-of-3 9.9 GB: 2-of-3 7.2 GB: 3-of-3 0.0 GB
Supported Languages	ko
Model Architecture	LlavaLlamaForCausalLM
License	apache-2.0
Context Length	2048
Model Max Length	2048
Transformers Version	4.28.0.dev0
Tokenizer Class	LlamaTokenizer
Vocabulary Size	32003
Torch Data Type	float32

Rank the KoLLaVA KoVicuna 7B 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 KoLLaVA KoVicuna 7B

Best Alternatives	Context / RAM	Downloads	Likes
Llava V1.6 Mistral 7B PATCHED	32K / 15.1 GB	17	8
Table Llava V1.5 7B	4K / 14.2 GB	9	13
Quilt Llava V1.5 7B	4K / 14.2 GB	610	9
Co Instruct Llava V1.5 7B	4K / 14.1 GB	6	1
...ct4V LLaVA Instruct Mix880k 7B	4K / 14.2 GB	7	3
...V1.5 7b Qinstruct Preview V0.1	4K / 14.2 GB	76	5
Chinese LLaVA Baichuan	2K / 28 GB	7	9
Llava 1.6 Gptq 8bit	32K / 9.6 GB	5	0

Note: green Score (e.g. "73.2") means that the model is better than tabtoyou/KoLLaVA-KoVicuna-7b.