ORAgentBench - Easy Feasibility: leaderboard

Metric: Mean feasibility (0-100) on the 32 easy tasks: share of submissions that satisfy the schema, domains and hard operational constraints; 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 2029. 14 models tracked.

Top models

#ModelScore
1GPT-5.4 (High)69
2Claude Opus 4.666
3DeepSeek V4 Flash62
4GPT-5.4 Mini (High)62
5MiMo-V2.5-Pro59
6Qwen 3.5 Plus59
7DeepSeek V4 Pro55
8GLM-5.153
9GLM-553
10Kimi K2.650
11Qwen 3.6 Plus50
12MiniMax-M2.750
13GPT-5.3 Codex (High)50
14Claude Sonnet 4.647

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