Dev Radar
Support
LiveUpdated 2026-11-03 23:11 UTC

Diffusion models are interesting because of decode speed.

Diffusion models are interesting because of decode speed. Now this new 8B model is giving me another reason to be…

This is a dev post classified by AGK Intelligence as Other (a launch), kept by the Dev Radar because it carries real work, not commentary.

Diffusion models are interesting because of decode speed. Now this new 8B model is giving me another reason to be excited about it, better math results for an 8B model. LLaCO-8B we released from @BAAIBeijing, BUPT and PKU. The researchers report these MATH scores... 📊 Previous continuous model DSL-LLaDA-8B: 15.10 📊 LLaDA-8B: 43.82 📊 Dream-7B: 51.20 🧠 LLaCO-8B: 62.12 Their approach changes how the model learns to recover text from noise, concentrating effort around the stages where token identities actually change. For us on Local AI, there’s something concrete to explore further 💾 Downlo

Posted by David Hendrickson (12.7k followers) 1 h ago · 3 likes · 327 views · view the original post on X. Kept by the Dev Radar as Other.

More dev work like this

Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.3k posts from 5k X accounts over the last 21 days, 9.2k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-03 23:11 UTC. Full method.