Market-Bench - Bid Efficiency: leaderboard

Market-Bench (2026) places LLM-driven retailer agents together in one simulated supply-chain market where they bid for stock in budget-constrained auctions, set prices and write slogans for persona-driven buyers; values are means over 10 runs of a 6-step simulation at temperature 0. This board is the won share of the quantity bid times the base value of the won goods over the price paid; higher is better.

Metric: Bid efficiency (won share of bid quantity times base value of won goods over the amount paid; 0 or more); 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 April 2027. 19 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 2.5 Flash0.795#466
2Gemini 2.5 Pro0.792#280
3O30.699#242
4GPT-4o0.45#754
5Market-Bench InternLM2.5 20B (checkpoint unspecified)0.441
6ERNIE 4.5 300B A47B0.418#1130
7Claude Sonnet 4.50.411#267
8Phi-40.386#1951
9Qwen 2.5 VL 72B Instruct0.318#849
10Market-Bench Gemma 3 27B (checkpoint unspecified)0.296
11InternLM3-8B-Instruct0.28#2714
12DeepSeek V3.20.237#425
13ERNIE 4.5 21B A3B0.216#2022
14Qwen 2.5 VL 32B Instruct0.197#1065
15Market-Bench Llama 3.1 70B (checkpoint unspecified)0.195

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=market-bench-bid-efficiency · How It Works · Data refreshed daily, snapshot 2026-11-22.