Have something to say?

Tell Moonlight how they could make the product more useful to you.

Offer to provide the option to choose a model for using generative AI

Hi, I'm a Moonlight user trying to keep up with the AI trends that are evolving almost as fast as Moonlight is evolving :) Your articles are always very helpful. I really appreciate it. It may be a simple perception on my part, but, looking at the quality of summarization generation for papers, it feels like the LLM API has recently changed from a different API service to a Gemini-based API. (I personally feel that the clarity and consistency of the summarized information has suffered somewhat). First off, I appreciate the developers' efforts to provide the best options as they evolve the product, Based on my experience, which could be a simple misunderstanding, I would like to suggest that you consider offering multiple choices for the LLM API. Nowadays, there are many different inference models available as API services, including o1, o3-mini, Gemini family, DeepSeek-R1, etc, I think a feature that allows users to choose a model based on their subscription tier would provide a better user experience and allow us to serve more users with a reasonable ROI. Also, I would like to cautiously suggest that you could also consider utilizing it in terms of performance marketing by identifying the most selected and preferred models. Thank you.

백성훈_코어팀(CF) ‍ 2 months ago