Meta Releases Llama 4 Scout: New Open Model with 1M Token Context
Meta has launched Llama 4 Scout, a new open-weights model featuring a native context window of 1 million tokens. The model utilizes a Mixture-of-Depths architecture for high computational efficiency.
What happened?
Meta and its research division, FAIR, have unveiled Llama 4 Scout, a new open-weights foundation model. The model is built on a dynamic Mixture-of-Depths (MoD) architecture and offers a native context window of 1 million tokens. In total, the model contains 105 billion parameters, but activates only 24 billion parameters per token.
Key facts
| Totalt antal parametrar | 105 miljarder |
|---|---|
| Aktiva parametrar per token | 24 miljarder |
| Kontextfönster | 1 miljon tokens |
| Lanseringsdatum | 11 oktober 2026 |
Why it matters
By utilizing Mixture-of-Depths, computational resources are dynamically allocated based on the complexity of a specific token, which significantly reduces inference costs. This allows the model to perform at the level of leading closed reasoning models in synthetic coding and long-context analysis, but at a substantially lower computational cost.
Who is affected?
Developers, AI researchers, and companies building large-scale systems for coding, mathematics, and long-document analysis are directly impacted. Because the weights are open, organizations can run the model on their own hardware or in private cloud environments without dependency on closed APIs.
Impact on the EU
Llama 4 Scout is released as an open-weights model, but as with previous Llama releases, its deployment in the EU may be subject to local privacy and AI regulations, such as the EU AI Act and GDPR. However, companies within the EU can run and fine-tune the model on their own infrastructure, provided that licensing conditions are met.
What else you should know
The model has a total parameter capacity of 105 billion, but activates only 24 billion per token due to MoD routing. The source indicates that the launch was made by Meta's AI research division, FAIR, on the DEV Community on 11 October 2026.
Quick answers about this story
Vad har hänt?
När hände det?
Varför spelar det roll?
Vilka berörs av lanseringen?
The link opens in a new window and leads to the publisher's own site.
Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.
AI-verktyg i artikeln
Topics
Get similar news straight to your inbox
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.
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: New Open Model with 1M Token Co"