Who would prefer that we offer a full online service?
Added 2023-10-26 01:06:47 +0000 UTCWe're really happy with what we have. Options! Online and fast, offline and private! Top-of-the-line or affordable! But... one of the thing we have heard from our members is that it's sometimes hard to get started if you're not aware of what those options mean, the kind of GPU you need to run a model locally, etc.
So we have a very important question for you today.
Should we offer an online service ourselves? One that would be specialized for Voxta, so you can just enter your key and get started?
We would design it with privacy first, try hard to make it affordable (cheaper than renting your own GPU, more expensive than slower paid services out there, unrestricted) and most of all, make it easy to use. No more ooba, no more options, just launch and go.
What do you think? Eventually this service could include TTS, STT and Text Gen but for now we'll focus on Text Gen / Action Inference services.
