Ollama chatbot deployment raises real cloud cost questions
A medical chatbot built with is running on a laptop through Ollama. Ollama runs the model on the developer’s own machine, so if only the app is deployed to AWS while the model stays on the laptop, public users would not normally be able to reach the model.
would likely require running the Ollama model on an internet-accessible server, such as AWS EC2 or another . That server may need more CPU, memory, or a GPU than a simple Flask app.
The main cost concern is whether hosting the model would make much more expensive. There is also a practical question about whether keeping the project local and listing it on a resume is acceptable for internship or placement interviews.
Key points
- The chatbot uses for a medical use case.
- Ollama was chosen because the model can run locally on a laptop.
- Deploying only the app to AWS would not make a laptop- available to public users.
- may require hosting Ollama on AWS EC2 or another .
- Model hosting can need more CPU, memory, or a GPU than a basic Flask app.