-
InfluxDB
Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.
Assuming we choose pipeline.ai's services, we would have to pay $0.00055 per second of GPU usage. If we assume we will have 4000 users messaging 50 times a day, and every inference would take 10 seconds, we're looking at ~$33,000 every month for inference costs alone. This is a very rough estimation, as the real number of users will very likely be much higher when a website launches, and it will be greater than 50 messages per day for each user. A more realistic estimate would put us at over $100k-$150k a month.
If its going to be AGPL that means you are allowed to use AI Hordes backend code if you wish. You can find it here : https://github.com/db0/AI-Horde/tree/KAI_DO_NOT_DELETE
Related posts
-
Nvidia Announces Financial Results for Second Quarter Fiscal 2024
-
Can u still run SD on google colab / kaggle free tier?
-
Running SD on external GPU?
-
Do you have to go to the colab and reinstall Pygmalion every time? And is it taking up space on my computer?
-
T-Shirt print designs model, made as a test. It requires zero prompt complexity. Model download link in comments