Llama-3.1-70B-Instruct-FP8

Name: Llama-3.1-70B-Instruct-FP8
Author: nvidia

70B open-weight LLM by nvidia

Live

Open / InstallLast updated August 5, 2026

Description

Llama-3.1-70B-Instruct-FP8 is an open-weight text-generation model with roughly 70B parameters published by nvidia on the Hugging Face Hub. It has 44,740 downloads. Runs with transformers.

Author

nvidia

Platform

api

Pricing model

free

Capabilities

Text generation
Conversational
Library: transformers
70B parameters

Benchmarks

No benchmark results have been added yet.