FDE-Bench - Health Monitoring: leaderboard

Metric: Resolved rate (%; the 34 health-monitoring-axis tasks (probes, metrics and alerts that must fire under injected faults and stay silent otherwise); 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
1Gemini 3.6 Flash82.4
2GPT-5.6 Terra79.4
3Claude Sonnet 5 (High)79.4
4DeepSeek V4 Pro76.5
5GPT-5.6 Luna73.5
6DeepSeek V4 Flash70.6
7Gemini 3.5 Flash Lite50

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