CodeRouterBench - LLM Router (0-Shot): leaderboard

Metric: Average task score (%) of the backend the model routes each task to, over the 2,919 in-distribution CodeRouterBench coding tasks (execution-graded or proxy scores of eight candidate LLMs, a missing result counting 0); the router picks one of the eight by name; zero-shot router prompt). Source: arxiv.org. Saturation forecast: Around 2031. 8 models tracked.

Top models

#ModelScore
1Qwen 3.5 Plus46.87
2Qwen 3 Max46.6
3Kimi K2.546.29
4GPT-5.445.91
5GLM-545.42
6MiniMax-M2.742.11
7Claude Sonnet 4.641.41
8Claude Opus 4.639.27

Interactive version: theaggregate.ai/benchmark?slug=coderouterbench-llm-router-0-shot · How It Works · Data refreshed daily, snapshot 2026-09-26.