1 個回答
- 最新
- 最多得票
- 最多評論
1
If you're solely using On-Demand inferencing, you'll only be charged for the tokens that you pass in and receive back from Bedrock as specified in the pricing page. However, If you decide to go the Provisioned Throughput route, you'll be charged for Model Units (includes a limited about of tokens) 24/7 for either a 1 month or 6 month terms.
已回答 3 個月前
相關內容
- 已提問 6 個月前
the "pricing page" link is wrong (goes to Credly)
Page here: https://aws.amazon.com/bedrock/pricing/