Georges Harik's humans& releases open-source 4-bit reinforcement learning recipe for long-horizon multi-agent training
The methodology resolves training instabilities associated with NVFP4-RL.
Combined views
73.1K
19 Sources, first seen ago
Sources
NI 
Niloofar ✈️ icml@niloofar_mireWe have open-sourced our 4-bit hardware native RL recipe, w/ NO degradation compared to 8 bit!! Checkout our blogpost! https://twitter.com/humansand/status/2075618383631167692
- likes: 181
- replies: 4
- bookmarks: 98
- reposts: 16
AC Andrew Carr 🤸@andrew_n_carrWhen you read a closed source blog pose, and they say "RL systems at scale take lots of engineer work", this is exactly the sort of thing they mean. Really great work by the sand team to go through the hard science and engineering effort to get low precision RL working AND go…
- likes: 46
- replies: 1
- bookmarks: 28
- reposts: 5
EL elie@eliebakouchi'm sad this is not happening near 6/7 of the range instead https://twitter.com/humansand/status/2075618383631167692
- likes: 221
- replies: 8
- bookmarks: 144
- reposts: 7
BZ Banghua Zhu@BanghuaZCongrats @humansand on native NVFP4 training recipe, landing in Miles! https://twitter.com/humansand/status/2075618383631167692
- likes: 71
- replies: 0
- bookmarks: 14
- reposts: 9
AP Andi Peng@TheAndiPenguinProud of the excellent work coming out of our infra team - excited for continued open source collab opportunities! https://twitter.com/humansand/status/2075618383631167692
- likes: 42
- replies: 0
- bookmarks: 15
- reposts: 1
CH Costa Huang@vwxyzjn- likes: 25
- replies: 2
- bookmarks: 4
- reposts: 0