What are the hardware requirements for Japanese Stablelm 3B 4e1t Instruct?

Japanese Stablelm 3B 4e1t Instruct requires approximately 5.6 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Japanese Stablelm 3B 4e1t Instruct and how large is it?

Japanese Stablelm 3B 4e1t Instruct is developed by stabilityai, a model with 3b 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 Stablelm 3B 4e1t Instruct?

Japanese Stablelm 3B 4e1t Instruct is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Japanese Stablelm 3B 4e1t Instruct by stabilityai — VRAM 5.6GB, 4K context

Name: Japanese Stablelm 3B 4e1t Instruct
Author: stabilityai

Japanese Stablelm 3B 4e1t Instruct is an open-source language model by stabilityai. Features: 3b LLM, VRAM: 5.6GB, Context: 4K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.09.

Arxiv:1607.06450 Arxiv:1910.07467 Arxiv:2104.09864 Arxiv:2204.06745 Arxiv:2307.09288 Custom code Instruct Ja Japanese-stablelm Region:us Safetensors Stablelm epoch

Model Card on HF 🤗: https://huggingface.co/stabilityai/japanese-stablelm-3b-4e1t-instruct

Japanese Stablelm 3B 4e1t Instruct Parameters and Internals

Model Type

text-generation, causal-lm

Use Cases

Areas:

Research, Commercial applications

Primary Use Cases:

As a foundational model for application-specific fine-tuning

Limitations:

Potential for generating offensive or inappropriate content.

Considerations:

Exercise caution in production systems to prevent harm or distress.

Supported Languages

Japanese (native)

Training Details

Data Sources:

Japanese translation of the Databricks Dolly-15k dataset, Japanese translation of the subset of the Anthropic HH dataset, Wikinews subset of the izumi-lab/llm-japanese-dataset

Methodology:

Fine-tuned on instruction-following datasets.

Context Length:

4096

Model Architecture:

Decoder-only transformer with modifications similar to LLaMA.

LLM Name	Japanese Stablelm 3B 4e1t Instruct
Repository 🤗	https://huggingface.co/stabilityai/japanese-stablelm-3b-4e1t-instruct
Model Size	3b
Required VRAM	5.6 GB
Updated	2026-08-07
Maintainer	stabilityai
Model Type	stablelm_epoch
Instruction-Based	Yes
Model Files	5.6 GB
Supported Languages	ja
Model Architecture	StableLMEpochForCausalLM
License	apache-2.0
Context Length	4096
Model Max Length	4096
Transformers Version	4.34.0
Tokenizer Class	GPTNeoXTokenizer
Vocabulary Size	50304
Torch Data Type	bfloat16

Rank the Japanese Stablelm 3B 4e1t Instruct 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 Japanese Stablelm 3B 4e1t Instruct

Best Alternatives	Context / RAM	Downloads	Likes
StableLM 3B EssayWriter	4K / 5.6 GB	20	3
Stablelm Zephyr 3B GPTQ	4K / 1.8 GB	10	12

Note: green Score (e.g. "73.2") means that the model is better than stabilityai/japanese-stablelm-3b-4e1t-instruct.