Llama 4 Scout

Open-weights MoE (17B active, 16 experts) with an industry-leading 10M token context window. Host-dependent pricing.

Llama 4 Scout strengths

• Industry-leading 10M context
• Open weights / self-hostable
• Very low cost
• Natively multimodal

Pricing & context

Cost guide: a typical call of ~10K input + 2K output tokens runs roughly ~$0.08 (varies by host) × 0.01 + ~$0.30 (varies by host) × 0.002 — worth modelling against cheaper tiers before committing high-volume traffic.

When to choose Llama 4 Scout

Llama 4 Scout is best for Extreme long-context tasks (whole codebases, large document sets) on an open, cheap model. If your workload is more cost-sensitive, weigh it against DeepSeek V4 Flash ($0.14 ($0.0028 cache hit) input /1M) first.

Llama 4 Scout FAQ

How much does Llama 4 Scout cost?

Llama 4 Scout is priced at ~$0.08 (varies by host) per 1M input tokens and ~$0.30 (varies by host) per 1M output tokens (public API list price), with a 10M tokens context window.

What is Llama 4 Scout best for?

Llama 4 Scout by Meta is best for Extreme long-context tasks (whole codebases, large document sets) on an open, cheap model.

Is Llama 4 Scout multimodal?

Llama 4 Scout supports text, image.

Tools that use Llama 4 Scout