What are the hardware requirements for KONI 4B Instruct 20250901?

KONI 4B Instruct 20250901 requires approximately 7.8 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed KONI 4B Instruct 20250901 and how large is it?

KONI 4B Instruct 20250901 is developed by KISTI-KONI, a model with 4b 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 KONI 4B Instruct 20250901?

KONI 4B Instruct 20250901 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

KONI 4B Instruct 20250901 by KISTI-KONI — VRAM 7.8GB, 128K context

Name: KONI 4B Instruct 20250901
Author: KISTI-KONI

KONI 4B Instruct 20250901 is an open-source language model by KISTI-KONI. Features: 4b LLM, VRAM: 7.8GB, Context: 128K, License: gemma, Instruction-Based, LLM Explorer Score: 0.2.

4b Base model:finetune:google/gem... Base model:google/gemma-3-4b-p... Conversational En Endpoints compatible Gemma3 Gemma3 text Instruct Ko Pytorch Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/KISTI-KONI/KONI-4B-instruct-20250901

KONI 4B Instruct 20250901 Parameters and Internals

LLM Name	KONI 4B Instruct 20250901
Repository 🤗	https://huggingface.co/KISTI-KONI/KONI-4B-instruct-20250901
Base Model(s)	Gemma 3 4B Pt google/gemma-3-4b-pt
Model Size	4b
Required VRAM	7.8 GB
Updated	2026-07-28
Maintainer	KISTI-KONI
Model Type	gemma3_text
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-2 2.8 GB: 2-of-2 15.5 GB 0.0 GB 0.0 GB
Supported Languages	ko en
Model Architecture	Gemma3ForCausalLM
License	gemma
Context Length	131072
Model Max Length	131072
Transformers Version	4.51.3
Tokenizer Class	GemmaTokenizer
Padding Token	<pad>
Vocabulary Size	262208
Torch Data Type	bfloat16

Rank the KONI 4B Instruct 20250901 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 KONI 4B Instruct 20250901

Best Alternatives	Context / RAM	Downloads	Likes
Gemma 3 4B T1 It	128K / 9.1 GB	27	41

Note: green Score (e.g. "73.2") means that the model is better than KISTI-KONI/KONI-4B-instruct-20250901.