All models
Meta: Llama 4 Scout
by Metatext
Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B.
- Est. AWS input
- US$0.17/M
- Est. AWS output
- US$0.66/M
- Context
- 1.3M tokens
- Max output
- 8k tokens
- Released
- Apr 2025
Estimated cost in your AWS account: AWS list price per million tokens. AWS bills you directly; your actual rate depends on your AWS agreement and region.
Prices synced from AWS · last updated
Providers
| Provider | Est. AWS input | Est. AWS output |
|---|---|---|
| Amazon Bedrock | US$0.17/M | US$0.66/M |
Govern spend on this model
Set budgets and alerts for every team that uses it, and see its cost by feature and customer.