Skip to content

cuda: GDN conv-chain fusion and alpha/beta pair at speculative verify widths (1-16 tokens, rollback snapshots), bit-identical, +5-6% MTP3 / +4-5% DFlash7 on H200 (GGML_CUDA_GDN_FUSION=1) - #399

Merged
bri-prism merged 1 commit into
prismfrom
cuda/gdn-fusion-verify
Oct 25, 2026
Merged

cuda: GDN conv-chain fusion and alpha/beta pair at speculative verify widths (1-16 tokens, rollback snapshots), bit-identical, +5-6% MTP3 / +4-5% DFlash7 on H200 (GGML_CUDA_GDN_FUSION=1)#399
bri-prism merged 1 commit into
prismfrom
cuda/gdn-fusion-verify

cuda: extend the GDN conv-chain fusion and alpha/beta mat-vec pair to…

5a65131
Select commit
Loading
Failed to load commit list.
Sign in for the full log view

Annotations

1 notice
Analyze (python)
succeeded Oct 25, 2026 in 58s