Open Systems, Blockchain & Business Daily

OSBBD

AI, money and the software behind them.

Llama 3.1 405B

Llama 3.1 405B is an open-weight language model from Meta, released on July 23, 2024.

Developer
Meta
Released
July 23, 2024
Total parameters
405B
Architecture
Dense
Context window
128K
Input
Text
License
Llama 3.1 Community License
Commercial use
Yes, with conditions
Official weights, GB
811.7 GB
Hugging Face repository
huggingface.co
Checked on

Llama 3.1 405B is a dense model with 405B parameters. Meta lists a context window of 128K tokens. The model takes text input. The weights are released under the Llama 3.1 Community License. Commercial use is allowed with conditions. Companies whose products had more than 700 million monthly active users need a separate license from Meta, and the Llama Acceptable Use Policy applies.

The official weights on Hugging Face take about 812 GB in BF16. Running the model needs at least that much memory across GPUs and system RAM, plus room for the context cache; quantized versions need less.

Total parameters, billions
405B
Context window, tokens
128K
License type
Custom license
License conditions
Companies whose products had more than 700 million monthly active users need a separate license from Meta, and the Llama Acceptable Use Policy applies.
Weights precision
BF16
Official quantized or alternative versions
meta-llama/Llama-3.1-405B-Instruct-FP8
License text
github.com
Sources
PageURL
Hugging Face model pagehttps://huggingface.co/meta-llama/Llama-3.1-405B-Instruct
Hugging Face APIhttps://huggingface.co/api/models/meta-llama/Llama-3.1-405B-Instruct?expand=safetensors
Hugging Face APIhttps://huggingface.co/api/models?author=meta-llama&search=FP8

Questions

What license does Llama 3.1 405B use?
The weights are released under the Llama 3.1 Community License.
Can I use Llama 3.1 405B commercially?
Yes, with conditions. Companies whose products had more than 700 million monthly active users need a separate license from Meta, and the Llama Acceptable Use Policy applies.
How much memory does Llama 3.1 405B need?
The official BF16 weights take about 812 GB; Meta also publishes an official FP8 version, which needs less. The memory calculator checks which setups fit.
How many parameters does Llama 3.1 405B have?
Llama 3.1 405B is a dense model with 405B parameters and a context window of 128K tokens.

Sources

Every figure comes from the developer’s model card, license text and official announcements, and from the model’s Hugging Face repository metadata. Each model page lists its sources and the date they were checked; the register is rechecked every week.

OSBBD

What are you looking for?

Search by headline, topic or keyword.