Skip to content

Qwen3.5-397B RL trainer on 128 v7x: fp8 fake-quant hoist, fused SwiGLU backward, attention backward savings, the wi forward GMM's weight pipeline depth and the GDN context-parallel m scan as config keys - #5799

Draft
parambole wants to merge 14 commits into
atwigg/mlperffrom
parambole/rl-4600-final-maxtext
Draft

Qwen3.5-397B RL trainer on 128 v7x: fp8 fake-quant hoist, fused SwiGLU backward, attention backward savings, the wi forward GMM's weight pipeline depth and the GDN context-parallel m scan as config keys#5799
parambole wants to merge 14 commits into
atwigg/mlperffrom
parambole/rl-4600-final-maxtext

Tests: gdn_cp_m_scan and gdn_cp_dh_scan, the GDN context-parallel sca…

1bf840c
Select commit
Loading
Failed to load commit list.
Google CLA / cla/google succeeded Oct 17, 2026 in 15s

✅ All contributors are covered under a CLA with Google

See https://cla.developers.google.com/ for more info about Google's Contributor License Agreement (CLA).

ℹ️ Googlers: Go here to view more details and manage scans for this pull request.

Details

The following contributors were found for this pull request:

✅ 1bf840c Author: @parambole <pa*****le​@google.com>
✅ 5fd3c95 Author: @khatwanimohit <moh*******ani​@google.com>

(Only the first commit for a unique contributor is listed.)

View more details on Google CLA