What are the hardware requirements for Japanese GPT Neox Small?

Japanese GPT Neox Small requires approximately 0.7 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 Japanese GPT Neox Small and how large is it?

Japanese GPT Neox Small is developed by rinna, a model with 203.6m 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 Japanese GPT Neox Small?

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

Japanese GPT Neox Small by rinna — VRAM 0.7GB, 2K context

Name: Japanese GPT Neox Small
Author: rinna

Japanese GPT Neox Small is an open-source language model by rinna. Features: 203.6m LLM, VRAM: 0.7GB, Context: 2K, License: mit, LLM Explorer Score: 0.1.

Arxiv:2101.00190 Arxiv:2404.01657 Dataset:cc100 Dataset:mc4 Dataset:wikipedia Gpt-neox Gpt neox Ja Lm Pytorch Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/rinna/japanese-gpt-neox-small

Japanese GPT Neox Small Parameters and Internals

Model Type

text generation, lm, nlp

Supported Languages

ja (native)

Training Details

Data Sources:

Japanese CC-100, Japanese C4, Japanese Wikipedia

Methodology:

Traditional language modelling objective

Model Architecture:

12-layer, 768-hidden-size transformer-based

Release Notes

Version:

2023/03/20

Date:

2023/03/20

Notes:

Update the model weight and config files to be compatible with Huggingface's official GPT-NeoX implementation.

LLM Name	Japanese GPT Neox Small
Repository 🤗	https://huggingface.co/rinna/japanese-gpt-neox-small
Model Size	203.6m
Required VRAM	0.7 GB
Updated	2026-07-31
Maintainer	rinna
Model Type	gpt_neox
Model Files	0.7 GB 0.7 GB
Supported Languages	ja
Model Architecture	GPTNeoXForCausalLM
License	mit
Context Length	2048
Model Max Length	2048
Tokenizer Class	T5Tokenizer
Padding Token	[PAD]
Vocabulary Size	44416
Torch Data Type	float32

Rank the Japanese GPT Neox Small 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