CodeRouterBench - LLM Router (3-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; three same-dimension probing examples with oracle labels and all eight scores). Source: arxiv.org. Saturation forecast: Around 2030. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GLM-5 | 46.03 |
| 2 | Qwen 3 Max | 45.8 |
| 3 | GPT-5.4 | 45.66 |
| 4 | Claude Sonnet 4.6 | 45.53 |
| 5 | Qwen 3.5 Plus | 45.39 |
| 6 | MiniMax-M2.7 | 45.29 |
| 7 | Kimi K2.5 | 45.1 |
| 8 | Claude Opus 4.6 | 41.17 |
Interactive version: theaggregate.ai/benchmark?slug=coderouterbench-llm-router-3-shot · How It Works · Data refreshed daily, snapshot 2026-09-26.