What are the hardware requirements for Taiwan ELM 270M?

Taiwan ELM 270M requires approximately 1.2 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Taiwan ELM 270M and how large is it?

Taiwan ELM 270M is developed by liswei, a model with 2b 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 Taiwan ELM 270M?

Taiwan ELM 270M is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Taiwan ELM 270M by liswei — VRAM 1.2GB

Name: Taiwan ELM 270M
Author: liswei

Taiwan ELM 270M is an open-source language model by liswei. Features: 2b LLM, VRAM: 1.2GB, License: apache-2.0, LLM Explorer Score: 0.12.

Base model:apple/openelm-270m Base model:finetune:apple/open... Conversational Custom code Dataset:liswei/taiwan-text-exc... Openelm Region:us Safetensors Tensorboard Zh

Model Card on HF 🤗: https://huggingface.co/liswei/Taiwan-ELM-270M

Taiwan ELM 270M Parameters and Internals

Model Type

text generation

Use Cases

Areas:

research

Considerations:

Model is designed for efficient operation on limited computational resources.

Additional Notes

The model is adaptable to different base models and data sizes depending on demand.

Supported Languages

zh (Traditional Chinese)

Training Details

Data Sources:

liswei/Taiwan-Text-Excellence-2B, liswei/PromptPair-TW

Data Volume:

2B Traditional Chinese tokens and 500K instruction samples

Methodology:

Custom fork of LLaMA-Factory

Input Output

Input Format:

~~[INST] <> {{ system_prompt }} <> {{ user_message }} [/INST]

Accepted Modalities:

text

LLM Name	Taiwan ELM 270M
Repository 🤗	https://huggingface.co/liswei/Taiwan-ELM-270M
Base Model(s)	liswei/Taiwan-ELM OpenELM 270M liswei/Taiwan-ELM apple/OpenELM-270M
Model Size	2b
Required VRAM	1.2 GB
Updated	2026-08-06
Maintainer	liswei
Model Type	openelm
Model Files	1.2 GB 0.0 GB
Supported Languages	zh
Model Architecture	OpenELMForCausalLM
License	apache-2.0
Transformers Version	4.40.1
Tokenizer Class	LlamaTokenizer
Padding Token	</s>
Vocabulary Size	61758
Torch Data Type	float32

Rank the Taiwan ELM 270M 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