Interface ModelTRTLLMRuntimeConfiguration

interface ModelTRTLLMRuntimeConfiguration {
    batch_scheduler_policy?: ModelTRTLLMBatchSchedulerPolicy;
    enable_chunked_context?: boolean;
    kv_cache_free_gpu_mem_fraction?: number;
    kv_cache_host_memory_bytes?: number | null;
    lora_cache_gpu_memory_fraction?: number | null;
    lora_cache_host_memory_bytes?: number | null;
    lora_cache_max_adapter_size?: number | null;
    lora_cache_optimal_adapter_size?: number | null;
    request_default_max_tokens?: number | null;
    served_model_name?: string | null;
    total_token_limit?: number;
    webserver_default_route?:
        | "/predict"
        | "/v1/embeddings"
        | "/rerank"
        | "/predict_tokens"
        | null;
    [k: string]: unknown;
}