Twitxy Saved

Yinan (Steven) Na

Verified account Creatify AI

@nyn531

CEO @Creatify_AI. Building AI that creates & optimizes ads. Previously Snap + Meta. Sharing what we learn building AI models and generating ads at scale.
Mountain View, California creatify.ai Joined March 2012
522 Following 2.9K Followers 5.7K Posts

Latest posts 5

Yinan (Steven) Na Verified account @nyn531
If you run video models on on-demand GPUs, this could cut your GPU bill in half. Cold starts go from over a minute to 7 seconds, so you stop paying for idle GPUs.

Really proud of the Creatify research team for open-sourcing this work.
Ledell Wu Verified account @LedellWu 4d
We’re open-sourcing Native Video Inference: the Rust + CUDA runtime we built at @Creatify_AI to serve video models.

Cold start to first denoising step: ~7s on a fresh B200, vs 65–75s on @sgl_project Diffusion and vLLM-Omni. Same output, bit for bit.

No PyTorch at inference. Kernels compiled ahead of time. Everything loads in parallel.

On our production traffic, that means 84% less idle GPU time.

Supports MiniMax H3, Wan 2.1 and LTX 2.5. Code and write-up below 👇
View quoted post on X
Yinan (Steven) Na Verified account @nyn531
Excited to launch a home base for @Creatify_AI Research, where we'll collect and share all of the work we're doing around model training, ads quality benchmark and research collaboration.
Creatify AI Verified account @Creatify_AI Oct 21
Introducing Creatify Labs.

Building better ad agents requires training specialized models, rigorous evaluation, benchmarks grounded in real ad production and campaign results, and research collaborations that push the frontier of video.

We’re sharing our research agenda and our work across each of these areas, starting with Boreal-H3, our video model trained for ads, and Aurora 2.0, our avatar model built for UGC ads.

Live at: labs.creatify.ai
View quoted post on X
23 replies 0 reposts 32 likes 1K views View on X