Qudata
Meta logo

Launch

Long-context multimodal MoE
By Metallama4:scout109B parameters32K tokens of contextLlama 4 Community License

Llama 4 Scout is a multimodal MoE with 109B total and 17B active parameters. It fits one 80GB datacentre GPU.

GPU configurations for Llama 4 Scout

Llama 4 Scout benchmarks

as published by the model vendor
Llama 4 Scout scores in public benchmarks
MMLU Pro74.3
GPQA Diamond57.2
LiveCodeBench32.8

Other models in the private loop

FAQ

Frequently asked about renting Llama 4 Scout

Which GPU does Llama 4 Scout need?

The entry configuration is 1x A100, which delivers around 28 tokens per second. For more parallel requests and headroom on speed, go with 1x H200.

How much does Llama 4 Scout cost?

An hour of Llama 4 Scout runs from $1.84 to $8.26 depending on the GPU configuration. Billing is hourly: you pay for server time, not for tokens.

How do I call Llama 4 Scout from my code?

Once deployed you get an OpenAI-compatible endpoint: point any OpenAI client at the base_url you receive and pass llama4:scout as the model name. Code written against the OpenAI API needs no changes.

Does the data stay private?

Yes. Llama 4 Scout runs on a dedicated GPU server inside a private loop: requests and responses never reach the model vendor and are not used for training.

What context window and license does Llama 4 Scout have?

Llama 4 Scout holds 32K tokens of context per request and ships under the Llama 4 Community License license, so you can use it commercially on your own hosting terms.