It was started by Shaw Walters, a personal friend and sometimes coworker.
The project is centered around a few core goals and a community building a larger ecosystem around the Eliza agentic framework.
It blew up because it caught a lot of hype at the right time. The framework itself is still working towards production-level maturity, but the community around it is certainly active.
The overall goal of the project was to create an on-chain agent that can manage its own assets and act as an investor which considers the input of key stakeholders of the token. Utility remains to be seen, but they are certainly heads down trying to build out their vision.
I can't speak for the average person involved in cryptocurrency because I have generally low opinions of the speculator market which completely shifted the original goals of Bitcoin away from real decentralization and towards centralized financial services in order to meet short-term goals. But certainly Shaw, I, and others in our circle are well aware of a16z and are generally tuned in to the financial and technopolitical landscape.
ai16z was renamed to elizaOS, you can view their website here: https://elizaos.ai/
Interesting business model. I was just speaking about app mascotification the other day with someone. It's a potential branch of the future. I wonder what the appetite is for curated avatars like this right now.
Regarding the tech... I really need to start offloading all of the tech I've built out in the last few years. I keep seeing business being erected on a fraction of the features I've developed in recent memory. Realtime lipsync, 3D avatars with generalized animation systems and basic autorigging, cost-effective/local TTS, STT, speaker diarization and embedding, memory models, agentic frameworks, there's just so much sitting on my drive in a handful of monorepos that needs proper treatment as either a standalone open source project or startup. I could take all this tech, grab a few artists I know and clone Mascotbot's business model in a week...
Not to detract from the great work they did here on coming up with and executing an idea. Subscription-based model work on its own is a really interesting market segment and I am very much rooting for Mascotbot's success.
Yeah, as I mentioned some of it is marked to be open sourced. Other bits are IP I'd like to retain but make actionable. Hopefully over the next few months I'll release some of it!
Thank you! I'm putting the finishing touches on a minimalist sh'mup that will play itself in the background. I've received your email and will reach out shortly :)
Hey all, just noticed someone featured us here! I'm Seraphim, the founder. Our mascot should have most info so feel free to chat on the website, but I'm happy to answer questions here as well
Great work! Brings a tear to my eye, as I had a fully operational system doing this in '08 with a tech team from Rhythm & Hues Studios, with two Oscar earners on the team too, and the investor class (all of them, seriously) insisted my startup do porn with the technology, and we refused. I think this time around you'll have better luck, but do watch out for the immature user population, they are large and demanding.
It's their property, and it can be integrated and use in their app after downgrading from design/animation-oriented plans. I see it as a normal situation when we worked on appearance and animations for several months, and then they just use it on our lower plans just using the API. The API will soon have more features supporting non-verbal speech-based animation/gestures, etc., so it will be even more valuable.
I can see one use already; Streamer avatar, but hosted instead of running it locally. All you would need to do is include the HTML in the OBS overlay and it can now be "your face" on stream.
I don't know of any vtubers who would want to spend $2490 then $200 a month to have an avatar instead of a one time payment of a lot less than that (depending on your artist/rigger). Corporate vtubers who would be able to afford that (like hololive) are able to get professionally made models, and indie ones are generally not rolling in the cash
Interesting but for $200/m and up I need to understand what exactly I am getting lover an open source project like ElizaOS/AI16z
What's the story behind Ai16z? I don't follow crypto, but that blew up from nowhere.
Who launched it? What are they doing? Why did they make so much money? Memes?
I didn't think the general crypto investing market knew about a16z.
It was started by Shaw Walters, a personal friend and sometimes coworker.
The project is centered around a few core goals and a community building a larger ecosystem around the Eliza agentic framework.
It blew up because it caught a lot of hype at the right time. The framework itself is still working towards production-level maturity, but the community around it is certainly active.
The overall goal of the project was to create an on-chain agent that can manage its own assets and act as an investor which considers the input of key stakeholders of the token. Utility remains to be seen, but they are certainly heads down trying to build out their vision.
I can't speak for the average person involved in cryptocurrency because I have generally low opinions of the speculator market which completely shifted the original goals of Bitcoin away from real decentralization and towards centralized financial services in order to meet short-term goals. But certainly Shaw, I, and others in our circle are well aware of a16z and are generally tuned in to the financial and technopolitical landscape.
ai16z was renamed to elizaOS, you can view their website here: https://elizaos.ai/
Interesting business model. I was just speaking about app mascotification the other day with someone. It's a potential branch of the future. I wonder what the appetite is for curated avatars like this right now.
Regarding the tech... I really need to start offloading all of the tech I've built out in the last few years. I keep seeing business being erected on a fraction of the features I've developed in recent memory. Realtime lipsync, 3D avatars with generalized animation systems and basic autorigging, cost-effective/local TTS, STT, speaker diarization and embedding, memory models, agentic frameworks, there's just so much sitting on my drive in a handful of monorepos that needs proper treatment as either a standalone open source project or startup. I could take all this tech, grab a few artists I know and clone Mascotbot's business model in a week...
Not to detract from the great work they did here on coming up with and executing an idea. Subscription-based model work on its own is a really interesting market segment and I am very much rooting for Mascotbot's success.
Perhaps open source it?
Yeah, as I mentioned some of it is marked to be open sourced. Other bits are IP I'd like to retain but make actionable. Hopefully over the next few months I'll release some of it!
Ha! love your co website.
Thank you! I'm putting the finishing touches on a minimalist sh'mup that will play itself in the background. I've received your email and will reach out shortly :)
Are the case studies fake? Can't find a trace of RPGLMS and the founder, Dillon Simmons appears to have an AI generated profile pic.
Not fake, some of these projects aren't published yet. You can see more of the work in progress on these and other projects on our Twitter/X
Hey all, just noticed someone featured us here! I'm Seraphim, the founder. Our mascot should have most info so feel free to chat on the website, but I'm happy to answer questions here as well
Great work! Brings a tear to my eye, as I had a fully operational system doing this in '08 with a tech team from Rhythm & Hues Studios, with two Oscar earners on the team too, and the investor class (all of them, seriously) insisted my startup do porn with the technology, and we refused. I think this time around you'll have better luck, but do watch out for the immature user population, they are large and demanding.
What happens to the art work if someone stops paying the membership? Is it their property or is it only available on your page?
Basically, who owns the likeness of the avatar and the artwork around it?
It's their property, and it can be integrated and use in their app after downgrading from design/animation-oriented plans. I see it as a normal situation when we worked on appearance and animations for several months, and then they just use it on our lower plans just using the API. The API will soon have more features supporting non-verbal speech-based animation/gestures, etc., so it will be even more valuable.
What exactly is the use case here? I have yet to see a website replace their text chat with an avatar
I can see one use already; Streamer avatar, but hosted instead of running it locally. All you would need to do is include the HTML in the OBS overlay and it can now be "your face" on stream.
I don't know of any vtubers who would want to spend $2490 then $200 a month to have an avatar instead of a one time payment of a lot less than that (depending on your artist/rigger). Corporate vtubers who would be able to afford that (like hololive) are able to get professionally made models, and indie ones are generally not rolling in the cash
We'll soon release our avatar builder and more affordable plans for those who need an easier start
I mean its hard to beat the pricing model of vroid studio + vseeface (both are free) or live2d (several free editors)
Looks really well executed. However I have my doubts about the business model.