Skip to content
Meta· LaunchAvailable

Meta Releases Llama 4 Scout and Maverick with 10M Context

Meta has released Llama 4 Scout and Maverick. These open models feature a 10-million-token context window and are built on a Mixture-of-Experts (MoE) architecture with 17 billion active parameters.

By the Aheadline editorial team·9 nov. 2026·2 min read·Source: Entity-watch: Meta AIVerifierad signalAI-generated
Meta Releases Llama 4 Scout and Maverick with 10M Context
Meta Releases Llama 4 Scout and Maverick with 10M Context
Meta Releases Llama 4 Scout and Maverick with 10M Context
By · Verktygs- & infrastrukturreporter
Last updated
Vad betyder det för mig?

What happened?

Meta has launched the AI models Llama 4 Scout and Llama 4 Maverick, which are based on a Mixture of Experts architecture with open model weights. Both models are multimodal from the ground up and feature a context window of up to 10 million tokens. Scout and Maverick each have 17 billion active parameters, distributed across 16 and 128 experts, respectively.

Key facts

Kontextfönster10 miljoner tokens
Aktiva parametrar17 miljarder (Scout och Maverick)
Antal experter i Scout16 experter
Antal experter i Maverick128 experter
Hårdvarukrav för ScoutRyms på en Nvidia H100 GPU

Why it matters

By offering open models with MoE architecture, computational costs during inference are reduced, as only a portion of the parameters are activated per token. This provides organisations with greater control over their data and lowers costs for tasks that require the handling of extensive context.

Who is affected?

The launch is aimed at AI developers, researchers, and companies looking to build multimodal applications with large amounts of data without relying on proprietary cloud services.

Impact on the EU

The models with open weights can be downloaded and run locally within the EU, enabling European organisations to manage data in compliance with local data protection requirements.

What else you should know

Scout is specifically optimised to fit onto a single Nvidia H100 graphics processor, which lowers the barrier for local deployment of models with extremely long context windows.

Frequently asked questions

Quick answers about this story

Vad har hänt?
Meta har släppt de öppna AI-modellerna Llama 4 Scout och Maverick med Mixture of Experts-arkitektur och 10 miljoner tokens i kontextfönster.
När hände det?
Modellerna presenterades i samband med Metas lansering av Llama 4-serien.
Varför spelar det roll?
Modellerna gör det möjligt att köra avancerade multimodala uppgifter och analysera enorma mängder text lokalt på standardiserad hårdvara utan slutna molntjänster.
Vad skiljer Scout från Maverick?
Scout har 16 experter och kan köras på en enda H100-GPU, medan Maverick har 128 experter för mer komplexa arbetsflöden.
Nyheten handlar om
MetaLlama 4 ScoutLlama 4 MaverickNvidia H100Mixture of Experts
Original source
Entity-watch: Meta AI·nownessai.com

The link opens in a new window and leads to the publisher's own site.

Verifierad signal

Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.

AI-verktyg i artikeln

Llama

Topics

#Meta#Multimodal AI#Models
[ STAY UP TO DATE ]

Get similar news straight to your inbox

No affiliate linksCancel anytimeGDPR-friendly
[ Frequency ]
[ What do you want to read about? ]

You'll receive updates on 2 topics.

The reader's room

Send in a question or an addition. The newsroom reads everything before it's published and replies when relevant. No AI-generated text – just people.

Sign in to submit a comment or question.

Loading comments…
How this affects you

Read the article through your role

  • Decide whether this affects strategy over 6–12 months or is just noise.
  • Discuss with leadership: do we own the right question or does ownership need to move?
  • Ask: what risk are we taking by NOT acting on this this quarter?

Generated angle — not editorial analysis of "Meta Releases Llama 4 Scout and Maverick with 10M Context"