/ Models Index / Llama 4 Scout

Llama 4 Scout

Open-Weights · Meta · Meta AI · Released Apr 2025
B

Good (B) — Capability Grade

Llama 4 Scout is Meta's smallest open-weights variant. Strong cost-efficiency; on-device deployment feasible. Mid-tier on most benchmarks.

80
Composite / 100
/ Subscore Breakdown · 6 Capability Dimensions

Where this grade comes from.

General Reasoning
B-
Code Generation
B-
Math & STEM
B-
Tool Use & Agency
C+
Multimodal
B
Safety & Alignment
B
/ Key Events & Disclosures

Release timeline & positioning.

  • Released Apr 2025
  • Smallest Llama 4 variant
  • Strong on-device feasibility
  • Active Llama 4 deployment

/ Best for

On-device AI applications; cost-efficient on-premise inference.

/ Watch out for

Frontier capability is materially lower than Behemoth or Maverick. Mid-tier reasoning.