๊ฐ€์ž… ํ›„ ์ดˆ๋Œ€ ๋งํฌ๋ฅผ ๊ณต์œ ํ•˜๋ฉด ๋™์˜์ƒ ์žฌ์ƒ ๋ฐ ์ดˆ๋Œ€ ๋ณด์ƒ์„ ๋ฐ›์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.

Ledell Wu
@LedellWu
Co-Founder & Chief Scientist @ Creatify AI ICML Test-of-Time Award Co-author of DPR (RAG-core) Past: FAIR @MetaAI
๊ฐ€์ž… May 2020
312 ํŒ”๋กœ์ž‰ ์ค‘    1.1K ํŒฌ
Weโ€™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 65โ€“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 ๐Ÿ‘‡
๋” ๋ณด๊ธฐ