Interface TRTLLMRuntimeConfigurationV2

interface TRTLLMRuntimeConfigurationV2 {
    enable_chunked_prefill?: boolean;
    max_batch_size?: number;
    max_num_tokens?: number;
    max_seq_len?: number | null;
    patch_kwargs?: {
        [k: string]:
            | string
            | number
            | unknown[]
            | { [k: string]: unknown }
            | null;
    };
    served_model_name?: string
    | null;
    tensor_parallel_size?: number;
    [k: string]: unknown;
}