All evidence cases
DeepSWE v1.1repositoryPython

Add safe import checkpoints and invariant validation

Add safe bulk import checkpoints, invariant validation, and rollback-on-failure behavior.

sqlite-utils-safe-import-checkpoints

Normalized profile

What this task asks for

Add safe bulk import checkpoints, invariant validation, and rollback-on-failure behavior. Create safe-import checkpoints that restore data and schema changes exactly on failure. Support nested checkpoint creation, commit, rollback, cleanup, and distinct inactive or unknown checkpoint errors. Persist table invariants and evaluate SELECT, aggregate, or row-level expressions with structured failure details. Add safe bulk insert, upsert, CSV import, and JSON import operations with strict and non-strict behavior. Expose safe-import and invariant management commands plus safe-mode options for insert, upsert, and bulk. Exit successfully only when safe-mode writes commit while invariant validation commands always exit zero.

  • feature implementation
  • configuration tuning
  • data transformation
  • test and validation
  • CLI
  • database schema
  • serialization
  • validation layer
  • library API
  • sqlite-utils
  • SQL
  • repository patch
  • strict behavioral compatibility or exact-output validation
  • stateful behavior and data-integrity constraints
  • specialized algorithmic or domain reasoning
  • changes span multiple components or interfaces

Published evidence

Model × harness outcomes

Counts are tied to these exact configurations. Harnesses are not treated as equivalent across benchmarks.

Models
13
Configurations
41
Trials
164

Showing 41 of 41 configurations

ModelHarnessEffortResultPasses
claude-fable-5vertex_aimini-swe-agentpublished configurationlowMixed3/4
claude-fable-5vertex_aimini-swe-agentpublished configurationmediumMixed2/4
claude-fable-5vertex_aimini-swe-agentpublished configurationhighAll passed4/4
claude-fable-5vertex_aimini-swe-agentpublished configurationxhighMixed3/4
claude-fable-5vertex_aimini-swe-agentpublished configurationmaxAll passed4/4
claude-opus-4-8anthropicmini-swe-agentpublished configurationlowMixed1/4
claude-opus-4-8anthropicmini-swe-agentpublished configurationmediumMixed3/4
claude-opus-4-8anthropicmini-swe-agentpublished configurationhighMixed2/4
claude-opus-4-8anthropicmini-swe-agentpublished configurationxhighMixed1/4
claude-opus-4-8anthropicmini-swe-agentpublished configurationmaxMixed3/4
claude-sonnet-4-6anthropicmini-swe-agentpublished configurationhighMixed1/4
claude-sonnet-5anthropicmini-swe-agentpublished configurationlowNo passes0/4
claude-sonnet-5anthropicmini-swe-agentpublished configurationmediumNo passes0/4
claude-sonnet-5anthropicmini-swe-agentpublished configurationhighNo passes0/4
claude-sonnet-5anthropicmini-swe-agentpublished configurationxhighNo passes0/4
claude-sonnet-5anthropicmini-swe-agentpublished configurationmaxNo passes0/4
gemini-3-1-pro-previewvertex_aimini-swe-agentpublished configurationhighNo passes0/4
gemini-3-5-flashvertex_aimini-swe-agentpublished configurationmediumMixed1/4
glm-5-2zaimini-swe-agentpublished configurationhighMixed1/4
glm-5-2zaimini-swe-agentpublished configurationmaxMixed1/4
gpt-5-4openaimini-swe-agentpublished configurationxhighMixed3/4
gpt-5-5openaimini-swe-agentpublished configurationlowMixed1/4
gpt-5-5openaimini-swe-agentpublished configurationmediumMixed2/4
gpt-5-5openaimini-swe-agentpublished configurationhighAll passed4/4
gpt-5-5openaimini-swe-agentpublished configurationxhighMixed3/4
gpt-5-6-lunaopenaimini-swe-agentpublished configurationlowNo passes0/4
gpt-5-6-lunaopenaimini-swe-agentpublished configurationmediumNo passes0/4
gpt-5-6-lunaopenaimini-swe-agentpublished configurationhighMixed3/4
gpt-5-6-lunaopenaimini-swe-agentpublished configurationxhighMixed3/4
gpt-5-6-lunaopenaimini-swe-agentpublished configurationmaxMixed3/4
gpt-5-6-solopenaimini-swe-agentpublished configurationlowMixed1/4
gpt-5-6-solopenaimini-swe-agentpublished configurationmediumMixed3/4
gpt-5-6-solopenaimini-swe-agentpublished configurationhighAll passed4/4
gpt-5-6-solopenaimini-swe-agentpublished configurationxhighAll passed4/4
gpt-5-6-solopenaimini-swe-agentpublished configurationmaxAll passed4/4
gpt-5-6-terraopenaimini-swe-agentpublished configurationlowMixed1/4
gpt-5-6-terraopenaimini-swe-agentpublished configurationmediumNo passes0/4
gpt-5-6-terraopenaimini-swe-agentpublished configurationhighMixed2/4
gpt-5-6-terraopenaimini-swe-agentpublished configurationxhighMixed3/4
gpt-5-6-terraopenaimini-swe-agentpublished configurationmaxAll passed4/4
kimi-k2-7-codeopenroutermini-swe-agentpublished configurationdefaultMixed1/4