Ask HN: Has anyone used DeepSeek in production?

11 pointsemgeee1y ago3 comments

There's been a lot of hype about the model for it's low inference costs and chain of thought reasoning but has anyone had actual success using it for production use-cases?

3 comments

marshughes1y ago

I also want to integrate DeepSeek's R1 model. I have to say that its capabilities are extremely powerful. I have compared the response levels of Claude 3.5 and the R1 model for the same questions, and there is a significant gap. For example, I am currently using Cursor and Windsurf for development simultaneously. After integrating the API of the R1 model, I feel that it has a slight edge in terms of planning and coding.

appidi_harinadh1y ago

Perplexity AI now allows you to ask queries with DeepSeek R1 if that's what you mean. I think AWS and Azure provides R1 in serverless mode so i guess there are companies that might be using in their own workflows.

gatsby12301y ago

Only with a small JXR wrapper. The API is much easier to control then. And yes, using it successfuly in production for user-made queries.

j / k navigate · click thread line to collapse