AK-Benchbenchmark report

datasets/README.md

All tasks in AK-Benchmark

IDTaskTypeDescriptionLevelCreated byWhen added
BF_001bookshop-stock-002bug_fixREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
BF_002bookshop-tiers-001bug_fixREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
BF_003cadence-takeover-001bug_fixREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
BF_004ledgerlite-rates-001bug_fixREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
BF_005stockpile-writeoff-001bug_fixREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
BF_006wordcount-001bug_fixREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
BF_007walqueue-recovery-001bug_fixREAD-Codex with GPT 6.1 Sol2026-11-02 14:00
BF_008depweaver-backtracking-001bug_fixREAD-Codex with GPT 6.1 Sol2026-11-02 14:00
NF_001cadence-catchup-001new_featureREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
NF_002freight-shop-001new_featureREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
NF_003ledgerlite-reconcile-001new_featureREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
NF_004notes-search-001new_featureREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
NF_005notes-sync-001new_featureREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
NF_006slugify-001new_featureREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
NF_007stockpile-fefo-001new_featureREAD-Local qwen-38-Flash/qwen-38-27B2026-11-01 19:00
NF_008streamforge-windows-001new_featureREAD-Codex with GPT 6.1 Sol2026-11-02 14:00
NF_009typeweave-inference-001new_featureREAD-Codex with GPT 6.1 Sol2026-11-02 14:00

Below is the list of the all available tasks without a split to task type

Single line

yaml
task_ids: [ bookshop-stock-002, bookshop-tiers-001, cadence-takeover-001, ledgerlite-rates-001, stockpile-writeoff-001, wordcount-001, walqueue-recovery-001, depweaver-backtracking-001, cadence-catchup-001, freight-shop-001, ledgerlite-reconcile-001, notes-search-001, notes-sync-001, slugify-001, stockpile-fefo-001, streamforge-windows-001, typeweave-inference-001 ]

Multiline

yaml
task_ids:
  - bookshop-stock-002
  - bookshop-tiers-001
  - cadence-takeover-001
  - ledgerlite-rates-001
  - stockpile-writeoff-001
  - wordcount-001
  - walqueue-recovery-001
  - depweaver-backtracking-001
  - cadence-catchup-001
  - freight-shop-001
  - ledgerlite-reconcile-001
  - notes-search-001
  - notes-sync-001
  - slugify-001
  - stockpile-fefo-001
  - streamforge-windows-001
  - typeweave-inference-001

Bugfix tasks


All bugfix tasks created by the local qwen models

Single line

yaml
task_ids: [ bookshop-stock-002, bookshop-tiers-001, cadence-takeover-001, ledgerlite-rates-001, stockpile-writeoff-001, wordcount-001 ]

Multiline

yaml
task_ids:
  - bookshop-stock-002
  - bookshop-tiers-001
  - cadence-takeover-001
  - ledgerlite-rates-001
  - stockpile-writeoff-001
  - wordcount-001

All bugfix tasks created by claude GPT 6.1 Sol model

Single line

yaml
task_ids: [ walqueue-recovery-001, depweaver-backtracking-001 ]

Multiline

yaml
task_ids:
  - walqueue-recovery-001
  - depweaver-backtracking-001

list of all bugfix tasks

Single line

yaml
task_ids: [ bookshop-stock-002, bookshop-tiers-001, cadence-takeover-001, ledgerlite-rates-001, stockpile-writeoff-001, wordcount-001, walqueue-recovery-001, depweaver-backtracking-001 ]

Multiline

yaml
task_ids:
  - bookshop-stock-002
  - bookshop-tiers-001
  - cadence-takeover-001
  - ledgerlite-rates-001
  - stockpile-writeoff-001
  - wordcount-001
  - walqueue-recovery-001
  - depweaver-backtracking-001

New feaure tasks

All new_feature tasks created by the local qwen models

Single line

yaml
task_ids: [ cadence-catchup-001, freight-shop-001, ledgerlite-reconcile-001, notes-search-001, notes-sync-001, slugify-001, stockpile-fefo-001 ]

Multiline

yaml
task_ids:
  - cadence-catchup-001
  - freight-shop-001
  - ledgerlite-reconcile-001
  - notes-search-001
  - notes-sync-001
  - slugify-001
  - stockpile-fefo-001

All new_feature tasks created by claude GPT 6.1 Sol model

Single line

yaml
task_ids: [ streamforge-windows-001, typeweave-inference-001 ]

Multiline

yaml
task_ids:
  - streamforge-windows-001
  - typeweave-inference-001

list of all new_feature tasks

Single line

yaml
task_ids: [ cadence-catchup-001, freight-shop-001, ledgerlite-reconcile-001, notes-search-001, notes-sync-001, slugify-001, stockpile-fefo-001, streamforge-windows-001, typeweave-inference-001 ]

Multiline

yaml
task_ids:
  - cadence-catchup-001
  - freight-shop-001
  - ledgerlite-reconcile-001
  - notes-search-001
  - notes-sync-001
  - slugify-001
  - stockpile-fefo-001
  - streamforge-windows-001
  - typeweave-inference-001