M

Mobillama 05B

Developed by MBZUAI
MobiLlama-05B is a small language model (SLM) with 500 million parameters, focusing on applications for resource-constrained devices, providing efficient and low-memory text generation capabilities.
Downloads 187
Release Time : 2/2/2024

Model Overview

MobiLlama-05B is a small language model designed based on the LLaMA-7B architecture, aiming to provide efficient and low-memory text generation solutions for resource-constrained devices. Its training data comes from the Amber dataset and supports English text generation tasks.

Model Features

Resource-efficient
Designed for resource-constrained devices, reducing pre-training and deployment costs, suitable for on-device processing and high-energy-efficiency scenarios.
Parameter-sharing scheme
Adopts a refined parameter-sharing scheme, significantly reducing the number of model parameters while maintaining performance.
Fully transparent open-source
Provides complete training data processes, training code, model weights, and evaluation code to ensure full transparency.

Model Capabilities

English text generation
Code generation
Technical Q&A

Use Cases

Privacy protection
On-device text generation
Generates text on local devices, avoiding data upload to the cloud and protecting user privacy.
Efficient computing
Low-memory applications
Suitable for devices with limited memory, such as mobile devices or embedded systems.
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase