Llama 3
Model ● Covered in 7 stories + Follow
Llama 3 is an open-weight large language model family released by Meta in 8B and 70B parameter sizes, with additional work noted for a 400B model still in training. Recent coverage highlights its role in the broader shift toward open source “frontier” model development, as well as its use in downstream tooling such as Llama-3-based VLM training workflows and agent frameworks. Articles also discuss performance-related findings connected to long-context behavior, where architectural factors studied in other models are reported alongside discussion of Llama 3’s ecosystem presence.
Updated 12 September 2026
Specifications
No specifications recorded yet.
Latest developments
April 2026
June 2025
May 2025
May 2024
April 2024
Relationships
Products & technology
- Meta develops this model · 2 sources
- Derived from Llama 2 · 1 source
- Integrated with Hugging Face · 1 source
- Integrated with Google Cloud · 1 source
- Integrated with Amazon SageMaker · 1 source