vLLM-SR configuration used in RouterArena evaluation

[truongdo619]

Thanks for open-sourcing RouterArena.

I’m currently trying to reproduce the reported vLLM-SR results in RouterArena, but I haven’t been able to identify the exact configuration used for the benchmark.

Would it be possible to share the following details for the vLLM-SR evaluation setup?

• The exact config.yaml file used to run vLLM-SR

If there were any additional settings or implementation details not included in the repository, I would also greatly appreciate clarification.

[jiarong0907]

Is this PR what you are looking for? #61

[truongdo619]

Thanks for the pointer, I took a look at PR #61.

I can see that vLLM-SR is run by calling the router endpoint at http://localhost:8080, but I’m still not fully sure how to serve and launch it correctly for reproduction.

Could you share the exact steps used to start the vLLM-SR router service (e.g., launch command, environment setup, required dependencies, or any additional configuration)?

[jiarong0907]

I believe you can find their router's choice of model for each query in the file called vllm-sr.json

[truongdo619]

Got it, thanks for the clarification.