AIM-Bench (inventory manager): leaderboard
AIM-Bench evaluates agentic LLMs acting as inventory managers across five uncertain supply-chain replenishment environments. It measures operational decision quality through inventory-management cost, stockout rate, and turnover rate, and diagnoses behavioral biases such as prospect-theory effects, mean anchoring, demand chasing, and bullwhip-effect dynamics. (Source: benchmarklist.com, self-repor
Metric: BG Avg. C (self-reported). Source: benchmarklist.com. Saturation forecast: Around October 2026. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V3 | 14.72 | #321 |
| 2 | GPT-4o | 15.14 | #343 |
| 3 | GPT-4.1 | 20.71 | #248 |
| 4 | Qwen 2.5 72B | 21.28 | #499 |
| 5 | Gemini 2.5 Flash Lite | 30.38 | #422 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=aim-bench-inventory-manager · How It Works · Data refreshed daily, snapshot 2026-10-20.