{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "title": "Info",
  "type": "object",
  "properties": {
    "model_id": {
      "type": "string",
      "description": "Loaded model ID"
    },
    "model_sha": {
      "type": "string",
      "description": "Model revision SHA"
    },
    "model_dtype": {
      "type": "string",
      "description": "Model data type"
    },
    "model_device_type": {
      "type": "string",
      "description": "Device type (cuda, cpu)"
    },
    "model_pipeline_tag": {
      "type": "string",
      "description": "Pipeline task tag"
    },
    "max_concurrent_requests": {
      "type": "integer"
    },
    "max_best_of": {
      "type": "integer"
    },
    "max_stop_sequences": {
      "type": "integer"
    },
    "max_input_length": {
      "type": "integer",
      "description": "Maximum input token length"
    },
    "max_total_tokens": {
      "type": "integer",
      "description": "Maximum total tokens (input + output)"
    },
    "waiting_served_ratio": {
      "type": "number"
    },
    "max_batch_total_tokens": {
      "type": "integer"
    },
    "max_waiting_tokens": {
      "type": "integer"
    },
    "validation_workers": {
      "type": "integer"
    },
    "max_client_batch_size": {
      "type": "integer"
    },
    "version": {
      "type": "string",
      "description": "TGI version"
    },
    "sha": {
      "type": "string",
      "description": "TGI build SHA"
    },
    "docker_label": {
      "type": "string",
      "description": "Docker image label"
    }
  }
}