Cofounder @HeyGen. Building @TryLiveAvatar and agents that talk, remember, and get work done.
Los Angeles, CA
- Tweets 717
- Following 955
- Followers 1,770
- Likes 1,612
Pinned Tweet
GPT-Live-1's voice + a real-time avatar = the language tutor I wish I had.
Full Japanese demo here. French, Korean, Spanish and more in the link below. Pick where you want to travel and start there 👇
Real-time AI experiences are now solved AND open sourced.
We worked closely with @OpenAI to build the best framework for it, combining GPT-Live-1, @TryLiveAvatar, and @HyperFrames_
Check out the demos below and show us what you build with it.
github.com/heygen-com/liveav…
17
52
749
152,594
realtime avatar stack, video bench, and voice clone for api. september was a shipping month.
In September, we shipped three tools for builders.
An open-source real-time avatar stack with OpenAI's GPT-Live-1
The Code2Video Benchmark with Kaggle, measuring if AI-made video is actually well made
And a Professional Voice Clone for API
Read More: heygen.com/blog/heygen-septe…
4
196
heygen video live on openrouter. distribution that fits how builders actually call models.
1/ @HeyGen Video 1 is now on OpenRouter.
Built for businesses. In blind side-by-side tests, it holds up against the top video models at a fraction of the price.
openrouter.ai/heygen/heygen-…
1
7
425
depth as the hierarchy is such a clean idea.
miguel keeps finding what sonnet can actually render.
all of this is code
Sonnet 5.5 max made this 39s video in 17 hours w/ @HyperFrames_ , prompt below
concept: depth is the hierarchy. On a screen you look into, focus picks what matters and going deeper shows more.
• One object: a black glass instrument with three windows, all showing the same tide reading, 3.67 m.
• Three depths, three levels: the shallowest window shows only the number, the middle one adds range, high water and rate, the deepest adds a trend and twelve readings.
• The camera pulls focus from window to window, so focus, not layout, decides what you read. It ends on the engraved line "Depth is the hierarchy".
• The design idea: information hierarchy carried by optical depth (glass, refraction, real depth of field) instead of size, weight and position on a flat screen. The starting point was Ferrari Luce's layered display.
• Built as one HyperFrames page: three.js glass and metal, live HTML text drawn into the 3D scene through HTML-in-Canvas, thin-lens depth of field, every frame deterministic.
3
132
Wayne L. retweeted
CPO of Ramp @geoffintech: I see 3 paths for product managers long-term:
1. The technical PM
2. The tastemaker PM
3. The GM
55
64
974
108,135
proud of the team shipping this.
production quality video shouldnt need a production budget.
We're releasing HeyGen Video, built for businesses that need production-quality video without production-level costs.
Pricing starts at $0.01/s through October (50% off)
Built on @Minimax_AI H3, post-trained by HeyGen.
Learn more: developers.heygen.com/heygen…
3
17
1,006
one prompt and 4.5h of sonnet cooking - this is what the stack was built for
this can sound crazy but this was done with only this prompt and sonnet 5.5 was left cooking 4.5h for making it:
"Make a hyperframes video, be as visually complex and premium and engaging as you can. Show off ALL of your visual and video making capabilities."
3
292
claude writing motion end to end - this is the unlock
I’ve been seeing a lot of people use Claude for coding, but this is a much more interesting use case.
Claude + HyperFrames can turn a simple idea into a finished motion graphics video.
Storyboard → animation → review → revisions.
That workflow could seriously change how freelancers approach motion design.
1
2
320
Opus 5.5 writes perfect HyperFrames_ videos
here's everything we learned so far from studying its model behavior
17
18
137
20,394
napoleon telling his own story - this is how education should feel
Napoleon POV explaining his rise and decline w/ Opus 5.5 & @HyperFrames_, this is the future of education
3
229
custom looks without losing the face - this is the fun part
And don't forget to dress for success.
Custom looks let you put your avatar in any outfit without losing what makes it look like you. We borrowed a look from Meta's Chief AI Officer @alexandr_wang, and you can too.
Same avatar from the weekly Slack update below, way more vibes.
2
3
301
slack to avatar video in four steps - this is how updates should work
New weekly update workflow
Step 1: open Claude Cowork
Step 2: "read this week's Slack and make me a video using the HeyGen MCP"
Step 3: pick from 3 scripts
Step 4: watch your avatar deliver it
Stop digging through Slack for what you missed. Watch the update instead
2
140
image to lego to motion - this loop is so clean
image -> lego -> @HyperFrames_ animation using @threejs
what have you tried with opus 5.5?
91
Wayne L. retweeted
This is a big enough problem that there's not just one startup idea here, but several. For example, there will probably be one variant meant to filter out agents, but another meant to welcome them in an orderly way. This also seems an idea you could build more on.
Bot detection & human verification will be one of the most urgent demands for businesses over the coming years. Agent swarms will suffocate every website and form; small companies and government websites are most vulnerable.
There is a huge gap in the market for this right now. When we looked at what offerings were in the market to use at X, there was not a single company that brought together all the latest technologies so we had to do it all in-house.
297
153
3,172
411,927