FDE-Bench - Repair: leaderboard

Metric: Resolved rate (%; the 70 repair tasks, which start from a deliberately broken deployment configuration; one episode per task, same four-tool scaffold (bash, write_file, read_file, submit), task-specific step, time and token caps; submitted artifacts are rebuilt and redeployed in a pristine sandbox and graded by four gated programmatic check layers (build, readiness, behavior, spec conformance); resolved only if all four pass). Source: arxiv.org. Saturation forecast: Around December 2026. 7 models tracked.

Top models

#ModelScore
1DeepSeek V4 Pro88.6
2DeepSeek V4 Flash88.6
3Gemini 3.6 Flash87.1
4Claude Sonnet 5 (High)87.1
5GPT-5.6 Terra80
6GPT-5.6 Luna74.3
7Gemini 3.5 Flash Lite68.6

Interactive version: theaggregate.ai/benchmark?slug=fde-bench-repair · How It Works · Data refreshed daily, snapshot 2026-09-26.