Llama 4 Scout

A natively multimodal Meta mixture-of-experts model: 17B active parameters out of 109B total, 16 experts, 10M token context. Accepts up to five images alongside text and fits on a single H100 when INT4-quantized. Twelve languages, though image understanding is stated for English only.

Understands

text, images

Produces

text, code

Contents verified with the vendor: 2026-08-18

Included in subscriptions

No catalogue entry marks this model as part of a plan yet. It will appear here once the plan contents are confirmed with the vendor.