4 ms·
When you have LLM requests you don't mind waiting for (up to 24h) then you can save 50% in costs. Great for document processing, image classification at scale,
by funerr 1y ago
When you have LLM requests you don't mind waiting for (up to 24h) then you can save 50% in costs. Great for document processing, image classification at scale, anything that you don't need an immediate result from the LLM provider and costs play a role.
- tomgs 1y agoConcrete use cases where 50 percent is actually a thing?