What are the hardware requirements for OpenELM 270M?

OpenELM 270M requires approximately 1.1 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed OpenELM 270M and how large is it?

OpenELM 270M is developed by apple, 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 OpenELM 270M?

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

OpenELM 270M by apple — VRAM 1.1GB

Name: OpenELM 270M
Author: apple

OpenELM 270M is an open-source language model by apple. Features: 270m LLM, VRAM: 1.1GB, License: apple-amlr, LLM Explorer Score: 0.14.

Arxiv:2404.14619 Custom code Openelm Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/apple/OpenELM-270M

OpenELM 270M Parameters and Internals

Model Type

Language Model

Additional Notes

Trained on publicly available datasets. Safety guarantees not provided.

Training Details

Data Sources:

RefinedWeb, deduplicated PILE, subset of RedPajama, subset of Dolma v1.6

Data Volume:

1.8 trillion tokens

Model Architecture:

Layer-wise scaling strategy

Input Output

Output Format:

Generated text based on input prompt.

Performance Tips:

To speed up inference, use speculative generation.

LLM Name	OpenELM 270M
Repository 🤗	https://huggingface.co/apple/OpenELM-270M
Model Size	270m
Required VRAM	1.1 GB
Updated	2026-08-08
Maintainer	apple
Model Type	openelm
Model Files	1.1 GB
Model Architecture	OpenELMForCausalLM
License	apple-amlr
Transformers Version	4.39.3
Vocabulary Size	32000
Torch Data Type	float32

Rank the OpenELM 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

Best Alternatives to OpenELM 270M

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
M2 DPO Oelm270m Fd	0K / 1.1 GB	9	0
...M 270M Instruct With Tokenizer	0K / 0.5 GB	6	0
Taiwan ELM 270M Instruct	0K / 1.2 GB	16	1
Oelm DPO 270m Fd	0K / 2.2 GB	8	0
DPO OpenELM 270M V1	0K / 0 GB	7	0
OpenELM 270M With Tokenizer	0K / 1.1 GB	7	0

Note: green Score (e.g. "73.2") means that the model is better than apple/OpenELM-270M.