@Keyuciallo

MSc @uni_tue; RA @MPI_IS

 Tübingen, Deutschland
keyu-wang-2002.github.io/
Joined November 2025
  • Tweets34
  • Following37
  • Followers210
Replying to @philna00
@philna00
2
36
DeepSeek
@deepseek_aiOct 20
🚀 Introducing DeepInfer, our in-house inference framework, now fully open source! Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable. 🔗 github.com/deepseek-ai/DeepI… 1/4

GitHub - deepseek-ai/DeepInfer: DeepInfer: High-performance, large-scale inference framework for...

DeepInfer: High-performance, large-scale inference framework for DeepSeek models - deepseek-ai/DeepInfer

github.com
1
4
215
Replying to @aHapBean
Great work (and xHC too)! We'd love to have your contributions! Feel free to open a PR to add to our collections, share any xHC/MSAR PR links in the Blogs section, or even join as a maintainer! keyu-wang-2002.github.io/Sca…
1
1
2
114
5/5 💡 Got a crazy idea? An unexpected result? An insight that didn't make it into your paper? Don't let it die in your notes! ScalingDepth welcomes blogs, research insights, and fresh perspectives from the community. Your ideas deserve to be seen! 🔥 Website: keyu-wang-2002.github.io/Sca…
1
3
159
4/5 Too many papers, too little time? We've got you! Explore research on residual connections, normalization, and depth scaling — all in one place. Stay curious. Stay updated. Beat the FOMO! 🚀
1
1
3
170
3/5 New architecture, new SOTA claim... but does it really scale with depth? Explore controlled comparisons from DepthBench, across different width–depth ratios at a fixed parameter budget. Less hype, more evidence! ⚖️
1
1
3
248
2/5 Confused by the ever-growing zoo of residual designs? We've brought them together in one gallery, with unified notation and intuitive formulas! One zoo. Countless ways to go deep.
1
1
4
699
mHC, iHC, oHC, xHC, and a zoo of AttnRes variants... An LLM architect with FOMO from the daily explosion of new papers? 🫠 Fear not! You can't afford to miss ScalingDepth — our community hub for keeping up with the ever-evolving world of residual connections! We'll keep tracking and sharing the latest research, methods, and papers from the community 🚀 And it's an OPEN community! Missed something? PRs and DMs are always welcome! Got a cool blog, post, idea, or unexpected finding? We'd love to hear it! ScalingDepth hosts a dedicated space to showcase community insights and discoveries.🌍 Built by the community. Driven by curiosity. Pushing the limits of depth! 🔥 🌐 keyu-wang-2002.github.io/Sca…
1
22
5
137
7,264
surprisingly, LDM is still not SOTA at extreme depth, which makes me even more convinced that depth scaling is a challenging yet promising direction!
1
1
286
super cool to see that Linear Depth Memory achieves impressive performance on DepthBench @_davidgonmar.
David González-Martínez@_davidgonmarOct 14
mHC and AttnRes demonstrate that richer depth mixing improves model performance at scale. Is there a unifying perspective? 🤔 Meet our Depth Memory framework: each layer reads from a set of streams to form its input, then writes back its output. Linear Depth Memory (LDM) equips models with a matrix-valued memory.
1
5
25
2,281
Load more