Market-Bench - Fill Rate: leaderboard
Metric: Fill rate (fraction of directed buyer demand in units that the agent fulfilled, 0 to 1); all 20 models compete as retailer agents in one simulated supply-chain market (sealed-bid procurement auctions, retail pricing and slogans; 6 steps, temperature 0), mean over 10 runs; higher is better. Source: arxiv.org. Saturation forecast: Around March 2027. 20 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O3 | 0.42 |
| 2 | Gemini 2.5 Pro | 0.22 |
| 3 | Gemini 2.5 Flash | 0.19 |
| 4 | Claude Sonnet 4.5 | 0.1 |
| 5 | Phi-4 | 0.09 |
| 6 | GPT-4o | 0.08 |
| 7 | Qwen 2.5 VL 72B Instruct | 0.05 |
| 8 | ERNIE 4.5 300B A47B | 0.05 |
| 9 | InternLM3-8B-Instruct | 0.03 |
| 10 | Qwen 2.5 VL 32B Instruct | 0.03 |
| 11 | DeepSeek V3.2 | 0.01 |
Interactive version: theaggregate.ai/benchmark?slug=market-bench-fill-rate · How It Works · Data refreshed daily, snapshot 2026-10-07.