Llama 3.1 405B

The Llama 3.1 405B model is a cutting-edge large language model released by Meta on July 23, 2024. Boasting 405 billion parameters, it utilizes an optimized transformer architecture with Grouped-Query Attention (GQA) for improved inference scalability. The model supports eight languages and is fine-tuned using supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF). Trained on approximately 15 trillion tokens with a knowledge cutoff in December 2023, it's designed for commercial and research applications, particularly in assistant-like chat scenarios. The model is available under a Community License, allowing for responsible use and modification .

Information

Family:
Llama 3.1
Released:
July 23, 2024
Parameters:
405B
Context:
128K
Variant:
base
Expert:
None
Repository:
HuggingFace
Knowledge Cutoff: