WebCoderBench: leaderboard

Metric: Overall score: preference-weighted sum of metric z-scores (% of a standard deviation). Source: arxiv.org. Saturation forecast: Around January 2027. 12 models tracked.

Top models

#ModelScore
1GPT-5 (High)13.81
2GLM-4.511.76
3MiniMax-M27.89
4DeepSeek V3.1 Terminus6.26
5DeepSeek R1 05282.41
6Gemini 2.5 Pro0.27
7GPT-5 Codex (High)-0.03
8Qwen 3 235B A22B Instruct-0.04
9Qwen 3 Coder Plus-2.31
10DeepSeek V3.1 (Thinking)-3.73
11DeepSeek V3.1-3.88
12GPT-4o (2024-11-20)-38.65

Interactive version: theaggregate.ai/benchmark?slug=webcoderbench · How It Works · Data refreshed daily, snapshot 2026-09-25.