Llama 3

library active open-source

Meta's open-weight large language model family released in 2024. Llama 3 models are available in 8B and 70B parameter sizes with competitive performance to proprietary models. The open weights enable local deployment, fine-tuning, and customization, making Llama 3 foundational to the open-source LLM ecosystem.

Implements

Concepts this tool claims to implement:

  • Large Language Model primary

    Competitive performance with proprietary models across benchmarks. Available in 8B and 70B parameter versions.

  • Foundation Model primary

    Base and instruction-tuned variants enable fine-tuning for specialized applications.

  • Chat Completions API secondary

    Instruction-tuned variants support chat completions format through various serving frameworks.

Integration Surfaces

  • Hugging Face
  • vLLM
  • Ollama
  • llama.cpp

Details

Vendor
Meta
License
Llama 3 Community License
Runs On
local, cloud
Used By
human, agent, system

Links

Notes

Llama 3 demonstrated that open-weight models can approach proprietary model performance. The open license (with some restrictions) enabled an ecosystem of fine-tunes, serving solutions, and applications built on Llama.

Related Tools