LLaMAX/LLaMAX3-8B
TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:8kPublished:Jun 25, 2024License:mitArchitecture:Transformer0.0K Open Weights Cold

LLaMAX/LLaMAX3-8B is an 8 billion parameter multilingual language base model, developed by Lu, Zhu, Li, Qiao, and Yuan through continued pre-training on Llama3. It supports over 100 languages, significantly enhancing translation capabilities for both high and low-resource languages. This model is designed to serve as a robust multilingual foundation for downstream tasks, excelling in translation performance.

Loading preview...