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

dots studio
@dotsstudioai
rednote's AI lab, creating frontier intelligence for daily life.
๊ฐ€์ž… August 2026
14 ํŒ”๋กœ์ž‰ ์ค‘    2.6K ํŒฌ
Under the hood of dots3-note Preview: BigMac ๐Ÿš€ One piece of the training infra behind dots3-note Preview is BigMac, our pipeline scheduling system purpose-built for VLM training. BigMac co-schedules heterogeneous vision and language workloads through a dependency-safe nested pipeline โ€” preserving the efficiency of optimized LLM schedules while reducing pipeline bubbles and activation memory. Across our evaluated workloads, it delivers 1.08โ€“1.9ร— training speedups over existing baselines. Weโ€™ve now open-sourced it.
๋” ๋ณด๊ธฐ