The flagship, 70 billion parameter language model from Meta, fine tuned for chat completions. Llama 2 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align to human preferences for helpfulness and safety.
Modalities
Context
4K
Released
Jun 20, 2023
Knowledge Cutoff
Jul 2023
The flagship, 70 billion parameter language model from Meta, fine tuned for chat completions. Llama 2 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align to human preferences for helpfulness and safety.
Llama 2 70B Chat has a 4,096 token context window.
Llama Guard 4 12B, Llama 4 Maverick, Llama 4 Scout and 5 more are other text models from Meta Llama.
Llama 2 70B Chat was released on June 20, 2023. Its knowledge cutoff is July 31, 2023.
Token volume and request traffic to this model over time.