Python Type-Check

Workflow runs for Python Type-Check

• ci: TAARDIS CUDA release workflow -- fat static CUDA 12.8 binaries to…

  cancelled:
  Python Type-Check#12:Commit32030c5pushedbyCodeMasterCody3D
  cuda-latest

  Sep 4, 20261d 0h 0m 2s
  More actions for workflow run 12

• Promote Q1_T_g128 to a full weight type + AVX2 kernel

  cancelled:
  Python Type-Check#11:Commit9fcc562pushedbyCodeMasterCody3D
  q1_0_g128-port

  Sep 2, 20261d 0h 0m 2s
  More actions for workflow run 11

• Forge rotation: golden tests, converter KV table, loader guard, graph…

  cancelled:
  Python Type-Check#10:Commit3519899pushedbyCodeMasterCody3D
  q1_0_g128-port

  Sep 1, 20261d 0h 0m 2s
  More actions for workflow run 10

• Port Q1_0_g128 + AVX2 kernel to upstream b356fa2 (enum 41 -> 43)

  cancelled:
  Python Type-Check#9:Commit495702bpushedbyCodeMasterCody3D
  q1_0_g128-port

  Sep 1, 20261d 0h 0m 11s
  More actions for workflow run 9

• convert: cite upstream for the Qwen3.8 tokenizer hash (verified, not …

  failed:
  Python Type-Check#8:Commit2e0caa2pushedbyCodeMasterCody3D
  q1_0_g128-avx2

  Sep 1, 20266m 32s
  More actions for workflow run 8

• convert: fix the Qwen3.8-27B tokenizer entry — right hash, and qwen35…

  failed:
  Python Type-Check#7:Commit2277a44pushedbyCodeMasterCody3D
  q1_0_g128-avx2

  Sep 1, 20266m 25s
  More actions for workflow run 7

• convert: register the Qwen3.8-27B tokenizer hash as the qwen2 pre-tok…

  failed:
  Python Type-Check#6:Commit3e53986pushedbyCodeMasterCody3D
  q1_0_g128-avx2

  Sep 1, 20263m 59s
  More actions for workflow run 6

• Q1_0_g128: full CPU support — quants, repack, AVX2 kernel, type regis…

  failed:
  Python Type-Check#5:Commit5aec1fdpushedbyCodeMasterCody3D
  q1_0_g128-avx2

  Sep 1, 20265m 10s
  More actions for workflow run 5

• Q1_0_g*: FP16 group scale (PrismML Q2_0 layout), replacing pow2 exponent

  failed:
  Python Type-Check#4:Commit10e988bpushedbyCodeMasterCody3D
  prism

  Aug 10, 20266m 10s
  More actions for workflow run 4

• Q1_0_g32 / g64 / g128: register three ternary types instead of one #d…

  failed:
  Python Type-Check#3:Commit3480262pushedbyCodeMasterCody3D
  prism

  Aug 10, 20262m 43s
  More actions for workflow run 3

• Q1_0_g128: ternary 2-bit with power-of-two scales, group 32

  failed:
  Python Type-Check#2:Commit4912378pushedbyCodeMasterCody3D
  prism

  Aug 10, 20262m 55s
  More actions for workflow run 2

• fix(vulkan): harden native TQ shader paths

  failed:
  Python Type-Check#1:Commitf92cac7pushedbyCodeMasterCody3D
  bonsai-vulkan-validation

  Jul 24, 202621h 55m 48s
  More actions for workflow run 1