What are the hardware requirements for MobileLLaMA 1.4B Base?

MobileLLaMA 1.4B Base requires approximately 2.7 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 MobileLLaMA 1.4B Base and how large is it?

MobileLLaMA 1.4B Base is developed by mtgv, a model with 1.4b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does MobileLLaMA 1.4B Base perform on standard benchmarks?

MobileLLaMA 1.4B Base has the following published scores: MMLU 24.79. Compare against reference models on this page or on the LLM Explorer leaderboards.

MobileLLaMA 1.4B Base by mtgv — VRAM 2.7GB, 2K context

Name: MobileLLaMA 1.4B Base
Rating: 4 (1 reviews)
Author: mtgv

MobileLLaMA 1.4B Base is an open-source language model by mtgv. Features: 1.4b LLM, VRAM: 2.7GB, Context: 2K, License: apache-2.0, LLM Explorer Score: 0.15, Arc: 34.4, HellaSwag: 56.3, MMLU: 24.8, GSM8K: 1.1.

Arxiv:2312.16886 Dataset:togethercomputer/redpa... Deploy:azure Endpoints compatible Llama Pytorch Region:us

Model Card on HF 🤗: https://huggingface.co/mtgv/MobileLLaMA-1.4B-Base

MobileLLaMA 1.4B Base Parameters and Internals

Model Type

Transformer

Additional Notes

MobileLLaMA aims to facilitate off-the-shelf deployment and reproducibility for further research.

Training Details

Data Sources:

RedPajama v1 dataset

Data Volume:

1.3T tokens

LLM Name	MobileLLaMA 1.4B Base
Repository 🤗	https://huggingface.co/mtgv/MobileLLaMA-1.4B-Base
Model Size	1.4b
Required VRAM	2.7 GB
Updated	2026-07-25
Maintainer	mtgv
Model Type	llama
Model Files	2.7 GB
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	2048
Model Max Length	2048
Transformers Version	4.33.1
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
Vocabulary Size	32000
Torch Data Type	float16

Rank the MobileLLaMA 1.4B Base 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

Quantized Models of the MobileLLaMA 1.4B Base

Model	Likes	Downloads	VRAM
MobileLLaMA 1.4B Base GGUF	3	207	0 GB

Best Alternatives to MobileLLaMA 1.4B Base

Best Alternatives	Context / RAM	Downloads	Likes
Llama 3.2 Hammered Four	128K / 2.8 GB	6	0
Llama3 1.4B Init 2	4K / 3 GB	4	1
42dot V0	4K / 5.8 GB	5	0
42dot Quantized	4K / 1.7 GB	5	0
42dot Number	4K / 5.8 GB	6	0
42dot Test Upload	4K / 5.8 GB	6	0
42dot Label	4K / 5.8 GB	5	0
42dot Law	4K / 5.8 GB	5	0
MobileLLaMA 1.4B Chat	2K / 5.5 GB	434347	21
ITT AF PLM 1.4B V0.1	2K / 5.8 GB	6	0

Note: green Score (e.g. "73.2") means that the model is better than mtgv/MobileLLaMA-1.4B-Base.