Meta-Llama-3.1-405B-Instruct
Multi language large-scale language model, optimizing dialogue scenarios.
- Supports 8 languages: English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai.
- Use Grouped Query Attention (GQA) technology to enhance inference scalability.
- Provide basic pre training models and instruction adjustment models, suitable for various natural language generation tasks.
- Follow responsible deployment strategies, including protecting developers and communities from potential abuse.
- Continuously improve model security through community feedback.
- Support commercial and research purposes, as well as using model outputs to enhance the capabilities of other models.
Product Details
Meta Llama 3.1 is a series of large-scale pre trained and instruction adjusted generative models in multiple languages, including versions of 8B, 70B, and 405B sizes. These models are optimized for multilingual conversation use cases and perform better than many open source and closed source chat models in common industry benchmarks. The model uses an optimized transformer architecture and is adjusted through supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to meet human preferences for usefulness and safety.
