config.pbtxt

backend: "ctranslate2"
name: "facebook_m2m100_1.2B"
max_batch_size: 64
input [
  {
    name: "INPUT_IDS"
    data_type: TYPE_INT32
    dims: [ -1 ]
    allow_ragged_batch: true
  }
]
input [
  {
    name: "TARGET_PREFIX"
    data_type: TYPE_INT32
    dims: [ -1 ]
    allow_ragged_batch: true
  }
]
output [
  {
    name: "OUTPUT_IDS"
    data_type: TYPE_INT32
    dims: [ -1 ]
  }
]
parameters [
  {
    key: "compute_type"
    value {
      string_value: "float16"
    }
  },
  {
    key: "max_decoding_length_multiple"
    value {
      string_value: "2"
    }
  },
    {
    key: "beam_size"
    value {
      string_value: "5"
    }
  }
]

instance_group [{ kind: KIND_GPU, count: 1 }]
dynamic_batching {
  max_queue_delay_microseconds: 5000
}