2 Risposte
- Più recenti
- Maggior numero di voti
- Maggior numero di commenti
2
Hi Joe, Based on the information provided, it seems your calculations for the estimated costs of using Amazon Bedrock's with RAG are correct. The key points are:
- For Titan Express : You estimated 4,000 input tokens and 27,000 output tokens per month. The aprox. pricing of only $0.05 per month for these usage levels.
- For Titan Text Embedding: You estimated a knowledge base of 40,000 tokens. The aprox. pricing of $0.005 per month.
It does seem that for the modest prototype you are considering, the costs would be effectively negligible. Your assumptions and calculations appear to be sound.
A few additional points to consider:
- The pricing for Amazon Bedrock is based on a pay-as-you-go model, so the costs will scale linearly as your usage increases.
- There are also charges for model customization, if you plan to create custom models beyond the pre-trained foundation models.
- It's always a good idea to monitor your actual usage and costs once you start using the service, as the pricing can change over time.
I suggest to reach out to AWS Support if you have any other questions about estimating or managing your costs further for the prototype.
con risposta un mese fa
Contenuto pertinente
- AWS UFFICIALEAggiornata 3 anni fa
- AWS UFFICIALEAggiornata un anno fa
- AWS UFFICIALEAggiornata 2 anni fa