What are the hardware requirements for Taiwan ELM 270M Instruct?

Taiwan ELM 270M Instruct 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 Instruct and how large is it?

Taiwan ELM 270M Instruct is developed by liswei, a model with 270m 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 Instruct?

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

Taiwan ELM 270M Instruct by liswei — VRAM 1.2GB

Name: Taiwan ELM 270M Instruct
Author: liswei

Taiwan ELM 270M Instruct is an open-source language model by liswei. Features: 270m LLM, VRAM: 1.2GB, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.13.

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

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

Taiwan ELM 270M Instruct Parameters and Internals

Model Type

text-generation

Use Cases

Areas:

Research

Applications:

Text Generation

Additional Notes

The model is notable for its efficiency, making it feasible for researchers with limited computing resources.

Supported Languages

zh (Traditional Chinese)

Training Details

Data Sources:

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

Data Volume:

2B Traditional Chinese tokens and 500K instruction samples

Methodology:

Custom fork of LLaMA-Factory

LLM Name	Taiwan ELM 270M Instruct
Repository 🤗	https://huggingface.co/liswei/Taiwan-ELM-270M-Instruct
Base Model(s)	Taiwan ELM 270M OpenELM 270M liswei/Taiwan-ELM-270M apple/OpenELM-270M
Model Size	270m
Required VRAM	1.2 GB
Updated	2026-07-27
Maintainer	liswei
Model Type	openelm
Instruction-Based	Yes
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 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 Taiwan ELM 270M Instruct

Best Alternatives	Context / RAM	Downloads	Likes
OpenELM 270M Instruct	0K / 0.5 GB	1255	145
OpenELM 270M Instruct	0K / 1.1 GB	8	0
...enELM270M DPO FINAL PREFERENCE	0K / 0.5 GB	6	0
M2 Oelm270m	0K / 1.1 GB	9	0
...M 270M Instruct With Tokenizer	0K / 0.5 GB	6	0
Mlnp Project Base	0K / 0.5 GB	5	0

Note: green Score (e.g. "73.2") means that the model is better than liswei/Taiwan-ELM-270M-Instruct.