If you watch the video you'll see that Aqua does more than just STT and why we need the frontier models. I think if you want just STT of what you said verbatim, then yea any local version of Whisper is great. If you want the enhancements shown in the video, you need the frontier models. And doing that locally is slow.
This is obviously a lie. If this was true, all the inference provider companies would go to zero. I support open-source as much as the next guy here, but it's obvious that the local version will be slower or break more often. Like, come on guys. Be real.
To illustrate this, M4 Max chips do 38 TOPS in FP8. An NVIDIA H100 does 4,000 TOPS.
Prakash if you're going to bot our replies, at least make it believable.
Users have different preferences on the text format they input into different apps. Aqua is able to pick up on these explicit and implicit preferences across apps – but no "open XYZ app" commands, yea
Hey thetopher, I'm the guy building the statue. My name is Pablo Peniche (@pablopeniche on twitter). I'm happy to put my name on it and answer questions.
The cost of the bigger-than-life-size statue (about 1.3x) is ~$25,000 (I'm happy to show you the receipts)
Shipping is ~$6,000 (it's "art" shipment so it's more expensive).
+ Installation at the park
+ Permits from city gov
+ Costs of the unveiling event
A 501c3 (not my own, but a tech-culture magazine "The Mars Review of Books") is sponsoring the project, so yea all donations are tax deductible.
I've personally paid ~$10,000 (again, I'm happy to show receipts) to get the design of the statue + website - and I made it all open-source so anyone can 3D print their own statue, for free.