we faced a problem of sharing various AI model subscriptions(like openAI, anthropic, gemini, etc., each costing $20/user/month) with our team of 10+ members, hence we decided to build a tool where we all can access 200+ AI models and collaborate on the AI tasks(like text and image generation, etc), it helps us save around 90% of our costs and saves a lot of time.
we shared this tool (flamingo.ai) with a few other teams and they love it and are willing to pay for it.
Are we on to something?!
Now we are really confused if we should go ahead and apply for Y combinator's fall'24 batch? help us decide!
thats a really good suggestion-We are thinking to make it more collaborative-sort of figma/slack where users can come and collab and share it.
but we want to grow super fast and we think YC can help us do that!
one of the things I would want in such custom client is that having features which are part of right their corresponding official clients like artifacts, python code execution, "whatever chatgpt does when the convo surpasses the context length" etc
Interesting concept, but I'm skeptical about the value of aggregating models vs. fine-tuning a few good ones. Have you considered offering fine-tuning services instead?
so thats exactly our plan going forward-once we see usage patterns of our user's, we will fine-tune models and also offer intelligent router to further optimize for speed and performance for the tasks.
8 comments
[ 2.0 ms ] story [ 34.8 ms ] threadAre we on to something?!
Now we are really confused if we should go ahead and apply for Y combinator's fall'24 batch? help us decide!
Create some sort of share / collab feature that will naturally spread to link minded users.