Home / Research Articles Hub / LLaMA: Open and Efficient Foundation Language Mode...
🤖 Artificial Intelligence OpenAlex

LLaMA: Open and Efficient Foundation Language Models

📅 Published: February 27, 2023 👤 Hugo Touvron, Thibaut Lavril, Gautier Izacard et al. 📖 arXiv (Cornell University) 📊 3,897 citations
AI-Generated Summary

We introduce LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. In particular, LLaMA-13B outperforms GPT-3 (175B) on most benchmarks, and LLaMA-65B is competitive with the best models, Chinchilla-70B and PaLM-540B.

⚡ This is an original paraphrased summary — not copied from the abstract. Full paper available at the source link below.

Key Findings
  • 1 We train our models on trillions of tokens, and show that it is possible to train advanced models using publicly available datasets exclusively, without resorting to proprietary and inaccessible datasets.
  • 2 In particular, LLaMA-13B outperforms GPT-3 (175B) on most benchmarks, and LLaMA-65B is competitive with the best models, Chinchilla-70B and PaLM-540B.
  • 3 We release all our models to the research community.
Why It Matters

This research advances how AI systems learn, reason, and solve problems — with direct implications for automation and scientific discovery.

This summary is based on publicly available metadata and abstract. For the full research paper, visit the original source:

Read Full Paper at OpenAlex
More Artificial Intelligence Papers ← Back to Hub 📚 Learning Hub
Article Details
Source OpenAlex
Category 🤖 Artificial Intelligence
Published Feb 27, 2023
Journal arXiv (Cornell University)
DOI 10.48550/arxiv.2302.13971
Citations 3,897
Authors Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux