0.70.2 T14: BLD-002 R8 known-failure ratchet against the last release (judges #4661-#4666) #4689

Activity

1. noahgift

   added

   P0Critical priorityCritical priority

   size:MK̂ 3-4 hK̂ 3-4 h

   owner:aprender-a7owner sessionowner session

   on Oct 4, 2026
2. noahgift

   added a parent issue on Oct 4, 2026

   • EPIC: release train 0.70.2 #4678
3. noahgift

   added sub-issues on Oct 4, 2026

   • 0.70.2 P0: Qwen2.5-0.5B-Instruct-f16.gguf cuda serve /api/chat: gibberish output (known failure in 0.70.1) #4661
   • 0.70.2 P0: Qwen3.5-0.8B-IQ4_XS.gguf golden_output thinking_on: measure the thinking budget for this model (known failure in 0.70.1) #4663
   • 0.70.2 P0: Qwen3.5-35B-A3B-UD-IQ4_XS.gguf capability_match: no CUDA forward for qwen35moe (known failure in 0.70.1) #4665
   • 0.70.2 P0: Qwen3-Coder-30B-A3B-Instruct-Q4_K_M.gguf cuda serve /api/chat: gibberish output (known failure in 0.70.1) #4662
   • 0.70.2 P0: Qwen3.5-0.8B-UD-IQ2_XXS.gguf golden_output: GPU forward falls back to CPU (known failure in 0.70.1) #4664
4. noahgift

   added 2 commits that reference this issue on Oct 4, 2026

   fix(release): known-failure ratchet review round 1 (Refs #4689)

   ...

   5adc7b9

   docs(lookahead): 0.70.2 ledger at 11:22Z: R8 and R9 built, three rows…

   ...

   36cfe0d
5. noahgift

   added

   owner:aprender-a7r8owning sessionowning session

   and removed

   owner:aprender-a7owner sessionowner session

   on Oct 5, 2026
6. noahgift

   added this to the 0.71.0 milestone on Oct 8, 2026
7. noahgift

   removed a parent issue on Oct 8, 2026

   • EPIC: release train 0.70.2 #4678
8. noahgift

   added a parent issue on Oct 8, 2026

   • EPIC E1 0.72: Fast Train #3998
9. noahgift

   added

   owner:aprender-w4664Owner session aprender-w4664Owner session aprender-w4664

   and removed

   owner:aprender-a7r8owning sessionowning session

   on Oct 8, 2026
10. noahgift

    modified the milestones: 0.71.0, 0.72.0 on Oct 9, 2026
11. noahgift commented on Oct 9, 2026

    

    noahgift

    on Oct 9, 2026

    ContributorAuthor

    More actions

    Parked under APR-EPIC-001 rule 5 (C334 budget cut; epic over its budget). No open PR or branch in the last 72 h. Reopen when it is staffed for its train.
12. noahgift

    closed this as not plannedon Oct 9, 2026
13. noahgift commented on Oct 9, 2026

    

    noahgift

    on Oct 9, 2026

    ContributorAuthor

    More actions

    C324 #6 bans a known-failure ratchet on a PR or release path. If this is wanted, it comes back as a nightly LAB row with one ticket and one owner.
14. noahgift

    removed sub-issues on Oct 14, 2026

    • 0.70.2 P0: Qwen2.5-0.5B-Instruct-f16.gguf cuda serve /api/chat: gibberish output (known failure in 0.70.1) #4661
    • 0.70.2 P0: Qwen3-Coder-30B-A3B-Instruct-Q4_K_M.gguf cuda serve /api/chat: gibberish output (known failure in 0.70.1) #4662
    • 0.70.2 P0: Qwen3.5-0.8B-IQ4_XS.gguf golden_output thinking_on: measure the thinking budget for this model (known failure in 0.70.1) #4663
    • 0.70.2 P0: Qwen3.5-0.8B-UD-IQ2_XXS.gguf golden_output: GPU forward falls back to CPU (known failure in 0.70.1) #4664
    • 0.70.2 P0: Qwen3.5-35B-A3B-UD-IQ4_XS.gguf capability_match: no CUDA forward for qwen35moe (known failure in 0.70.1) #4665