ORAgentBench - Hard Quality: leaderboard

Metric: Mean normalized objective quality (0-100) on the 34 hard tasks, zero for infeasible submissions and 0.5 at the verified reference solution; end-to-end operations research tasks packaged with multi-file operational data, configuration and a submission schema; the agent writes and runs solver code in an isolated Harbor environment (45-minute interaction budget, 5-minute solve limit) and hidden validators score schema and hard-constraint feasibility and normalized objective quality; printed 0-1 and x100 here; higher is better. Source: arxiv.org. Saturation forecast: Around April 2028. 14 models tracked.

Top models

#ModelScore
1GPT-5.4 (High)21
2GPT-5.3 Codex (High)19
3Claude Sonnet 4.614
4Claude Opus 4.614
5DeepSeek V4 Pro14
6Kimi K2.613
7GLM-5.110
8DeepSeek V4 Flash9
9Qwen 3.6 Plus7
10GPT-5.4 Mini (High)7
11GLM-56
12MiMo-V2.5-Pro4
13MiniMax-M2.73
14Qwen 3.5 Plus3

Interactive version: theaggregate.ai/benchmark?slug=oragentbench-hard-quality · How It Works · Data refreshed daily, snapshot 2026-09-29.