Touvron and colleagues: Llama 2 foundation and chat model family

What changed

Released and evaluated Llama 2 pretrained models from 7B to 70B parameters and fine-tuned dialogue variants, documenting training and safety methods.

Scientific scope

Released and evaluated Llama 2 pretrained models from 7B to 70B parameters and fine-tuned dialogue variants, documenting training and safety methods. Scope limitation: This is the 2023 Llama 2 research report, distinct from the original LLaMA and later Llama 3. Benchmarks and safety evaluations are those reported by the team.

Evidence

The retained primary-source abstract, bibliographic record and stated review scope support this provisional assessment. No independent experimental replication by WRS is claimed. This is the 2023 Llama 2 research report, distinct from the original LLaMA and later Llama 3. Benchmarks and safety evaluations are those reported by the team.

Contributors

• The named authors listed in the linked research record · SCIENCE: Complete source byline retained in the research record; human research direction, method development and evaluation. Individual labor shares are not measured.

• Touvron and colleagues: Llama 2 foundation and chat model family — original paper
• Touvron and colleagues: Llama 2 foundation and chat model family — company research page

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction