A community member reported encountering a TypeError with use_beam_search being an unexpected keyword argument while using vllm. Through discussion, it was revealed that this issue likely stems from a breaking change in a recent vllm version that removed certain parameters.
While the exact version couldn't be confirmed as the user was testing on run pod and had closed the kernel, another community member indicated that a patch would be needed and shared a link to a relevant pull request in the llama_index repository.
