OSS, Lambda, Bedrock costs

Hello.

utilisation (on-demand only) - "Data Instances will be always at 100% utilization dedicated for your Amazon Opensearch Service domain, to stop incurring charges you must Delete the domain." **-- will deleting the domain delete my chatbot? can my chatbot work if I delete the domain associated with it? **

Are you using Bedrock Knowledge Base for your chatbot app?
If you are using the Bedrock knowledge base, you will no longer be able to use it after removing Opensearch Serverless.

If you use a chatbot that uses your own RAG instead of the Bedrock knowledge base, I think you can use it even if you remove Opensearch Service depending on how the application is made.
I can't say anything about this since it depends on how the application is designed.
If you remove Opensearch Service, you will no longer be able to search your data sources using RAG, whether you are using the Bedrock Knowledge Base or not.

what instances are recommended to build a chatbot with about 10GB of data stored (ie the data files that will be used in the RAG process)?

As stated in the documentation below, there is no perfect method for sizing the OpenSearch Service.
The document below contains factors that can be used as a reference for sizing, so please use them to make an initial estimate.
https://docs.aws.amazon.com/opensearch-service/latest/developerguide/sizing-domains.html

There's no perfect method of sizing Amazon OpenSearch Service domains. However, by starting with an understanding of your storage needs, the service, and OpenSearch itself, you can make an educated initial estimate on your hardware needs. This estimate can serve as a useful starting point for the most critical aspect of sizing domains: testing them with representative workloads and monitoring their performance.

https://docs.aws.amazon.com/opensearch-service/latest/developerguide/sizing-domains.html#bp-instances

In general, the storage limits for each instance type map to the amount of CPU and memory that you might need for light workloads. For example, an m6g.large.search instance has a maximum EBS volume size of 512 GiB, 2 vCPU cores, and 8 GiB of memory. If your cluster has many shards, performs taxing aggregations, updates documents frequently, or processes a large number of queries, those resources might be insufficient for your needs. If your cluster falls into one of these categories, try starting with a configuration closer to 2 vCPU cores and 8 GiB of memory for every 100 GiB of your storage requirement.