tokenspeed / TensorFold SM80 support (PR 1302 closed won't fix; archive kernels may now include sm_80); NVIDIA flare-local-inference is Blackwell-only (no Ampere port)

  • Swimlane: Default swimlane
  • Column: Blocked
  • Position: 10
  • Assignee: not assigned
  • Creator: Ivan
  • Started:
  • Created: 10/30/2026 09:17
  • Modified: 10/30/2026 09:17
  • Moved: 10/30/2026 09:17
  • 8 GB
  • 10 GB
Description

Reported by / on it: @c133, @voidgiraffe, @dufrelic; blocked upstream

Source: https://github.com/lightseekorg/tokenspeed

Wiki page: Engines (Engines that do not run on sm_80)

Cards: both

Card D-081, seeded from the cmpunlocker Discord with credit (reference github:lightseekorg/tokenspeed).

Sub-Tasks
Internal links
Comments