Llama 3.1 405B Instruct FP8 by nvidia

 ยป  All LLMs  ยป  nvidia  ยป  Llama 3.1 405B Instruct FP8   URL Share it on

  Autotrain compatible Base model:finetune:meta-llama... Base model:meta-llama/llama-3....   Conversational   Endpoints compatible   Instruct   Llama   Region:us   Safetensors   Sharded   Tensorflow

Llama 3.1 405B Instruct FP8 Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Llama 3.1 405B Instruct FP8 (nvidia/Llama-3.1-405B-Instruct-FP8)

Llama 3.1 405B Instruct FP8 Parameters and Internals

LLM NameLlama 3.1 405B Instruct FP8
Repository ๐Ÿค—https://huggingface.co/nvidia/Llama-3.1-405B-Instruct-FP8 
Base Model(s)  meta-llama/Llama-3.1-405B-Instruct   meta-llama/Llama-3.1-405B-Instruct
Model Size405b
Required VRAM209.2 GB
Updated2025-05-17
Maintainernvidia
Model Typellama
Instruction-BasedYes
Model Files  4.8 GB: 1-of-86   4.9 GB: 2-of-86   4.6 GB: 3-of-86   4.9 GB: 4-of-86   4.6 GB: 5-of-86   4.9 GB: 6-of-86   4.6 GB: 7-of-86   4.9 GB: 8-of-86   4.6 GB: 9-of-86   4.9 GB: 10-of-86   4.6 GB: 11-of-86   4.9 GB: 12-of-86   4.6 GB: 13-of-86   4.9 GB: 14-of-86   4.6 GB: 15-of-86   4.9 GB: 16-of-86   4.6 GB: 17-of-86   4.9 GB: 18-of-86   4.6 GB: 19-of-86   4.9 GB: 20-of-86   4.6 GB: 21-of-86   4.9 GB: 22-of-86   4.6 GB: 23-of-86   4.9 GB: 24-of-86   4.6 GB: 25-of-86   4.9 GB: 26-of-86   4.6 GB: 27-of-86   4.9 GB: 28-of-86   4.6 GB: 29-of-86   4.9 GB: 30-of-86   4.6 GB: 31-of-86   4.9 GB: 32-of-86   4.6 GB: 33-of-86   4.9 GB: 34-of-86   4.6 GB: 35-of-86   4.9 GB: 36-of-86   4.6 GB: 37-of-86   4.9 GB: 38-of-86   4.6 GB: 39-of-86   4.9 GB: 40-of-86   4.6 GB: 41-of-86   4.9 GB: 42-of-86   4.6 GB: 43-of-86   4.9 GB: 44-of-86
Model ArchitectureLlamaForCausalLM
Licensellama3.1
Context Length131072
Model Max Length131072
Transformers Version4.44.0
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Llama 3.1 405B Instruct FP8

Best Alternatives
Context / RAM
Downloads
Likes
Meta Llama 3.1 405B Instruct128K / 186 GB55654473
Llama 3.1 405B Instruct128K / 183.1 GB29577569
...ta Llama 3.1 405B Instruct FP8128K / 197.6 GB55370165
Llama 3.1 405B Instruct FP8128K / 193.4 GB3594188
BigLlama 3.1 681B Instruct128K / 190.8 GB1911
INTELLECT 1 Instruct8K / 20.5 GB44122
...ama 3.1 405B Instruct Bnb 4bit128K / 214.3 GB406
Note: green Score (e.g. "73.2") means that the model is better than nvidia/Llama-3.1-405B-Instruct-FP8.

Rank the Llama 3.1 405B Instruct FP8 Capabilities

๐Ÿ†˜ Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! ๐ŸŒŸ

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 47402 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227