What are the hardware requirements for Kuririn?

Kuririn requires approximately 10.2 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Kuririn and how large is it?

Kuririn is developed by AtosArruda, a model with 5.1b 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 Kuririn?

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

Kuririn by AtosArruda — VRAM 10.2GB

Name: Kuririn
Author: AtosArruda

Kuririn is an open-source language model by AtosArruda. Features: 5.1b LLM, VRAM: 10.2GB, License: apache-2.0.

Base model:finetune:unsloth/ge... Base model:unsloth/gemma-4-e2b... Conversational Dataset:unsloth/latex ocr Endpoints compatible Gemma4 Image-text-to-text Latex Ocr Region:us Safetensors Sharded Tensorflow Unsloth Vision

Model Card on HF 🤗: https://huggingface.co/AtosArruda/Kuririn

Kuririn Parameters and Internals

LLM Name	Kuririn
Repository 🤗	https://huggingface.co/AtosArruda/Kuririn
Base Model(s)	Gemma 4 E2B It unsloth/gemma-4-E2B-it
Model Size	5.1b
Required VRAM	10.2 GB
Updated	2026-11-05
Maintainer	AtosArruda
Model Type	gemma4
Model Files	1.4 GB: 1-of-4 4.7 GB: 2-of-4 2.1 GB: 3-of-4 2.0 GB: 4-of-4
Model Architecture	Gemma4ForConditionalGeneration
License	apache-2.0
Model Max Length	131072
Transformers Version	5.5.0
Tokenizer Class	GemmaTokenizer
Padding Token	<pad>
Torch Data Type	bfloat16

Rank the Kuririn 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 Kuririn

Best Alternatives	Context / RAM	Downloads	Likes
Gemma 4 E2B It	0K / 10.2 GB	3222438	654
Gemma 4 E2B	0K / 10.2 GB	79938	425
Gemma 4 E2B It	0K / 10.2 GB	69653	13
Ringg Router E2b	0K / 10.2 GB	740	7
Gemma 4 E2B It Heretic	0K / 10.2 GB	1340	0
Gemma4 E2B ASR E2 BF16	0K / 10.2 GB	48	0
Gemma SEA LION V4.5 E2B IT	0K / 10.2 GB	2783	12
Heretic Scalpel E2B	0K / 10.2 GB	139	2
...ma 4 E2B Patent Claim Revision	0K / 10.2 GB	44	1
Gemma4 E2B FACEBOOK E2 BF16	0K / 10.2 GB	20	0

Note: green Score (e.g. "73.2") means that the model is better than AtosArruda/Kuririn.