feat(qwen4_exp): tensor parallelism for Qwen3.8-Flash-Next (offload backend) - #385

Draft

gdevenyi wants to merge 2 commits into

FlashML-org:mainfrom

gdevenyi:feat/qwen4-exp-tp

Commits on Sep 4, 2026

feat(qwen4_exp): tensor parallelism for Qwen3.8-Flash-Next (offload backend)

gdevenyi
and
claude
committed
9e73f2c
fix(qwen4_exp): load the HF config for sharding only when TP > 1

gdevenyi
and
claude
committed
d16d634