Llama 3.1 70B Instruct

Conditional

Meta · Llama · Released 2024-07

70BText

A mid-2024 workhorse dense model that still holds up for general reasoning and coding when 405B-class compute is not available.

Strengths

  • +Strong general reasoning for a dense 70B
  • +Wide third-party fine-tune and GGUF support
  • +128K context window

Limitations

  • -Needs 2+ high-VRAM GPUs or heavy quantization to self-host
  • -Surpassed on most benchmarks by Llama 3.3 70B

License

Llama 3.1 Community License

free with acceptable-use or scale conditions - read it

Hardware

wants 24-48GB of VRAM - a 3090/4090-class card or better

Links

WeightsGitHubPaper

Stats

meta-llama/Llama-3.1-70B-Instruct— downloads·— likes

via Hugging Face

Related models

Llama 3.1 8B Instruct

Meta · Llama

Llama 3.1 405B Instruct

Meta · Llama

Llama 3.2 3B Instruct

Meta · Llama

Llama 3.3 70B Instruct

Meta · Llama