SIGN IN SIGN UP

[AMD] added Minimax-M3-MXFP8 high concurrency notes on sparse-pa and INT4 quick-reduce quant (#654)

Signed-off-by: Hongxia Yang <hongxia.yang@amd.com>
H
Hongxia Yang committed
16898a579a86e979abceae4deb3bfb53e05c5327
Parent: ed141b1
Committed by GitHub <noreply@github.com> on 7/21/2026, 7:05:40 AM