A3M Router, an open-source LLM router with parallel multi-LLM execution, has achieved the #1 ranking on the RouterArena leaderboard — the first open-source project to top the benchmark, beating Microsoft Azure, OpenAI GPT-5, and every commercial routing service.
| Rank | Router | Score | Cost/1K | Type |
|---|---|---|---|---|
| 🥇 | A3M Router | 76.43 | $0.047 | Open-source |
| 🥈 | Sqwish | 75.27 | $0.18 | Closed-source |
| 🥉 | Azure-Model-Router (Microsoft) | 71.87 | $0.22 | Closed-source |
| 4 | R2-Router (UCF) | 71.60 | $0.06 | Open-source |
| 5 | GPT-5 (OpenAI) | 64.32 | $10.02 | Closed-source |
| 6 | NotDiamond | 57.29 | $4.10 | Closed-source |
| 7 | RouteLLM (UC Berkeley) | 48.07 | $0.27 | Open-source |
A3M is the highest-ranked and lowest-cost router on the leaderboard — $0.047/1K queries, 3.8x cheaper than the nearest competitor.
RouterArena (arXiv:2510.00202) is the standardized open benchmark for LLM routing systems. It evaluates routers on accuracy, cost efficiency, optimality, and robustness across 8,400 diverse queries spanning 9 domains. Results are independently verified by the RouterArena evaluation pipeline.
Unlike every other router on the leaderboard that uses sequential model selection (try one model, if it fails try the next), A3M runs providers simultaneously and scores responses by confidence — a technique called parallel ensemble execution. This is why it achieves the highest accuracy at the lowest cost.
npx a3m-router route "Your query here"
Or install globally:
npm install -g adaptive-memory-multi-model-routerView on GitHub → View on npm →
For media inquiries, open a discussion on GitHub Discussions or reach out via the repository.