LongRCA-Mini (All-at-Once): leaderboard

Metric: Root-step accuracy (%; exact match to the earliest decisive root-cause step; LongRCA-Mini, 200 fixed human-annotated failed agent trajectories, 40 from each of SWE-bench Pro, Terminal-Bench 2, TravelPlanner, VitaBench and WebArena Verified (macro-average of the five sources); All-at-once protocol: the whole execution log in one attribution prompt; invalid outputs count as errors). Source: arxiv.org. Saturation forecast: Around 2029. 5 models tracked.

Top models

#ModelScore
1GPT-5.6 Sol17
2DeepSeek V4 Pro11.5
3Kimi K311
4DeepSeek V4 Flash7.5
5Qwen 3.5 35B A3B3.5

Interactive version: theaggregate.ai/benchmark?slug=longrca-mini-all-at-once · How It Works · Data refreshed daily, snapshot 2026-09-29.