Stream responses from Generative AI models

Streaming involves receiving responses to prompts as they are generated. That is, as soon as the model generates output tokens, the output tokens are sent.

You can make streaming requests to the Vertex AI Large Language Model (LLM) using the following:

The streaming and non-streaming APIs use the same parameters, and there is no difference in pricing and quotas.

Vertex AI Studio

You can use Vertex AI Studio to design and run prompts and receive streamed responses. From the prompt design page, click the Streaming Response button to enable streaming.

Streaming response button

Supported languages

Language code Language
en English
es Spanish
pt Portuguese
fr French
it Italian
de German
ja Japanese
ko Korean
hi Hindi
zh Chinese
id Indonesian

Examples

You can call the Streaming API by using one of the following:

REST API with server-sent events (SSE)

The parameters are different across the model types used in the following examples:

Text

The current supported models are text-bison and text-unicorn. See available versions.

Request

  PROJECT_ID=YOUR_PROJECT_ID
  PROMPT="PROMPT"
  MODEL_ID=text-bison

  curl \
  -X POST \
  -H "Authorization: Bearer $(gcloud auth print-access-token)" \
  -H "Content-Type: application/json" \
  https://us-central1-aiplatform.googleapis.com/v1/projects/${PROJECT_ID}/locations/us-central1/publishers/google/models/${MODEL_ID}:serverStreamingPredict?alt=sse -d \
  '{
    "inputs": [
      {
        "struct_val": {
          "prompt": {
            "string_val": [ "'"${PROMPT}"'" ]
          }
        }
      }
    ],
    "parameters": {
      "struct_val": {
        "temperature": { "float_val": 0.8 },
        "maxOutputTokens": { "int_val": 1024 },
        "topK": { "int_val": 40 },
        "topP": { "float_val": 0.95 }
      }
    }
  }'

Response

Responses are server-sent event messages.

  data: {"outputs": [{"structVal": {"content": {"stringVal": [RESPONSE]},"safetyAttributes": {"structVal": {"blocked": {"boolVal": [BOOLEAN]},"categories": {"listVal": [{"stringVal": [Safety category name]}]},"scores": {"listVal": [{"doubleVal": [Safety category score]}]}}},"citationMetadata": {"structVal": {"citations": {}}}}}]}

Chat

The current supported model is chat-bison. See available versions.

Request

PROJECT_ID=YOUR_PROJECT_ID
PROMPT="PROMPT"
AUTHOR="USER"
MODEL_ID=chat-bison

curl \
-X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json" \
https://us-central1-aiplatform.googleapis.com/v1/projects/${PROJECT_ID}/locations/us-central1/publishers/google/models/${MODEL_ID}:serverStreamingPredict?alt=sse -d \
$'{
  "inputs": [
    {
      "struct_val": {
        "messages": {
          "list_val": [
            {
              "struct_val": {
                "content": {
                  "string_val": [ "'"${PROMPT}"'" ]
                },
                "author": {
                  "string_val": [ "'"${AUTHOR}"'"]
                }
              }
            }
          ]
        }
      }
    }
  ],
  "parameters": {
    "struct_val": {
      "temperature": { "float_val": 0.5 },
      "maxOutputTokens": { "int_val": 1024 },
      "topK": { "int_val": 40 },
      "topP": { "float_val": 0.95 }
    }
  }
}'

Response

Responses are server-sent event messages.

data: {"outputs": [{"structVal": {"candidates": {"listVal": [{"structVal": {"author": {"stringVal": [AUTHOR]},"content": {"stringVal": [RESPONSE]}}}]},"citationMetadata": {"listVal": [{"structVal": {"citations": {}}}]},"safetyAttributes": {"structVal": {"blocked": {"boolVal": [BOOLEAN]},"categories": {"listVal": [{"stringVal": [Safety category name]}]},"scores": {"listVal": [{"doubleVal": [Safety category score]}]}}}}}]}

Code

The current supported model is code-bison. See available versions.

Request

PROJECT_ID=YOUR_PROJECT_ID
PROMPT="PROMPT"
MODEL_ID=code-bison

curl \
-X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json" \
https://us-central1-aiplatform.googleapis.com/v1/projects/${PROJECT_ID}/locations/us-central1/publishers/google/models/${MODEL_ID}:serverStreamingPredict?alt=sse -d \
$'{
  "inputs": [
    {
      "struct_val": {
        "prefix": {
          "string_val": [ "'"${PROMPT}"'" ]
        }
      }
    }
  ],
  "parameters": {
    "struct_val": {
      "temperature": { "float_val": 0.8 },
      "maxOutputTokens": { "int_val": 1024 },
      "topK": { "int_val": 40 },
      "topP": { "float_val": 0.95 }
    }
  }
}'

Response

Responses are server-sent event messages.

data: {"outputs": [{"structVal": {"citationMetadata": {"structVal": {"citations": {}}},"safetyAttributes": {"structVal": {"blocked": {"boolVal": [BOOLEAN]},"categories": {"listVal": [{"stringVal": [Safety category name]}]},"scores": {"listVal": [{"doubleVal": [Safety category score]}]}}},"content": {"stringVal": [RESPONSE]}}}]}

Code chat

The current supported model is codechat-bison. See available versions.

Request

PROJECT_ID=YOUR_PROJECT_ID
PROMPT="PROMPT"
AUTHOR="USER"
MODEL_ID=codechat-bison

curl \
-X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json" \
https://us-central1-aiplatform.googleapis.com/v1/projects/${PROJECT_ID}/locations/us-central1/publishers/google/models/${MODEL_ID}:serverStreamingPredict?alt=sse -d \
$'{
  "inputs": [
    {
      "struct_val": {
        "messages": {
          "list_val": [
            {
              "struct_val": {
                "content": {
                  "string_val": [ "'"${PROMPT}"'" ]
                },
                "author": {
                  "string_val": [ "'"${AUTHOR}"'"]
                }
              }
            }
          ]
        }
      }
    }
  ],
  "parameters": {
    "struct_val": {
      "temperature": { "float_val": 0.5 },
      "maxOutputTokens": { "int_val": 1024 },
      "topK": { "int_val": 40 },
      "topP": { "float_val": 0.95 }
    }
  }
}'

Response

Responses are server-sent event messages.

data: {"outputs": [{"structVal": {"safetyAttributes": {"structVal": {"blocked": {"boolVal": [BOOLEAN]},"categories": {"listVal": [{"stringVal": [Safety category name]}]},"scores": {"listVal": [{"doubleVal": [Safety category score]}]}}},"citationMetadata": {"listVal": [{"structVal": {"citations": {}}}]},"candidates": {"listVal": [{"structVal": {"content": {"stringVal": [RESPONSE]},"author": {"stringVal": [AUTHOR]}}}]}}}]}

REST API

The parameters are different across the model types used in the following examples:

Text

The current supported models are text-bison and text-unicorn. See available versions.

Request

  PROJECT_ID=YOUR_PROJECT_ID
  PROMPT="PROMPT"
  MODEL_ID=text-bison

  curl \
  -X POST \
  -H "Authorization: Bearer $(gcloud auth print-access-token)" \
  -H "Content-Type: application/json" \
  https://us-central1-aiplatform.googleapis.com/v1/projects/${PROJECT_ID}/locations/us-central1/publishers/google/models/${MODEL_ID}:serverStreamingPredict -d \
  '{
    "inputs": [
      {
        "struct_val": {
          "prompt": {
            "string_val": [ "'"${PROMPT}"'" ]
          }
        }
      }
    ],
    "parameters": {
      "struct_val": {
        "temperature": { "float_val": 0.8 },
        "maxOutputTokens": { "int_val": 1024 },
        "topK": { "int_val": 40 },
        "topP": { "float_val": 0.95 }
      }
    }
  }'

Response

{
  "outputs": [
    {
      "structVal": {
        "citationMetadata": {
          "structVal": {
            "citations": {}
          }
        },
        "safetyAttributes": {
          "structVal": {
            "categories": {},
            "scores": {},
            "blocked": {
              "boolVal": [
                false
              ]
            }
          }
        },
        "content": {
          "stringVal": [
            RESPONSE
          ]
        }
      }
    }
  ]
}

Chat

The current supported model is chat-bison. See available versions.

Request

PROJECT_ID=YOUR_PROJECT_ID
PROMPT="PROMPT"
AUTHOR="USER"
MODEL_ID=chat-bison

curl \
-X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json" \
https://us-central1-aiplatform.googleapis.com/v1/projects/${PROJECT_ID}/locations/us-central1/publishers/google/models/${MODEL_ID}:serverStreamingPredict -d \
$'{
  "inputs": [
    {
      "struct_val": {
        "messages": {
          "list_val": [
            {
              "struct_val": {
                "content": {
                  "string_val": [ "'"${PROMPT}"'" ]
                },
                "author": {
                  "string_val": [ "'"${AUTHOR}"'"]
                }
              }
            }
          ]
        }
      }
    }
  ],
  "parameters": {
    "struct_val": {
      "temperature": { "float_val": 0.5 },
      "maxOutputTokens": { "int_val": 1024 },
      "topK": { "int_val": 40 },
      "topP": { "float_val": 0.95 }
    }
  }
}'

Response

{
  "outputs": [
    {
      "structVal": {
        "candidates": {
          "listVal": [
            {
              "structVal": {
                "content": {
                  "stringVal": [
                    RESPONSE
                  ]
                },
                "author": {
                  "stringVal": [
                    AUTHOR
                  ]
                }
              }
            }
          ]
        },
        "citationMetadata": {
          "listVal": [
            {
              "structVal": {
                "citations": {}
              }
            }
          ]
        },
        "safetyAttributes": {
          "listVal": [
            {
              "structVal": {
                "categories": {},
                "blocked": {
                  "boolVal": [
                    false
                  ]
                },
                "scores": {}
              }
            }
          ]
        }
      }
    }
  ]
}

Code

The current supported model is code-bison. See available versions.

Request

PROJECT_ID=YOUR_PROJECT_ID
PROMPT="PROMPT"
MODEL_ID=code-bison

curl \
-X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json" \
https://us-central1-aiplatform.googleapis.com/v1/projects/${PROJECT_ID}/locations/us-central1/publishers/google/models/${MODEL_ID}:serverStreamingPredict -d \
$'{
  "inputs": [
    {
      "struct_val": {
        "prefix": {
          "string_val": [ "'"${PROMPT}"'" ]
        }
      }
    }
  ],
  "parameters": {
    "struct_val": {
      "temperature": { "float_val": 0.8 },
      "maxOutputTokens": { "int_val": 1024 },
      "topK": { "int_val": 40 },
      "topP": { "float_val": 0.95 }
    }
  }
}'

Response

{
  "outputs": [
    {
      "structVal": {
        "safetyAttributes": {
          "structVal": {
            "categories": {},
            "scores": {},
            "blocked": {
              "boolVal": [
                false
              ]
            }
          }
        },
        "citationMetadata": {
          "structVal": {
            "citations": {}
          }
        },
        "content": {
          "stringVal": [
            RESPONSE
          ]
        }
      }
    }
  ]
}

Code chat

The current supported model is codechat-bison. See available versions.

Request

PROJECT_ID=YOUR_PROJECT_ID
PROMPT="PROMPT"
AUTHOR="USER"
MODEL_ID=codechat-bison

curl \
-X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json" \
https://us-central1-aiplatform.googleapis.com/v1/projects/${PROJECT_ID}/locations/us-central1/publishers/google/models/${MODEL_ID}:serverStreamingPredict -d \
$'{
  "inputs": [
    {
      "struct_val": {
        "messages": {
          "list_val": [
            {
              "struct_val": {
                "content": {
                  "string_val": [ "'"${PROMPT}"'" ]
                },
                "author": {
                  "string_val": [ "'"${AUTHOR}"'"]
                }
              }
            }
          ]
        }
      }
    }
  ],
  "parameters": {
    "struct_val": {
      "temperature": { "float_val": 0.5 },
      "maxOutputTokens": { "int_val": 1024 },
      "topK": { "int_val": 40 },
      "topP": { "float_val": 0.95 }
    }
  }
}'

Response

{
  "outputs": [
    {
      "structVal": {
        "candidates": {
          "listVal": [
            {
              "structVal": {
                "content": {
                  "stringVal": [
                    RESPONSE
                  ]
                },
                "author": {
                  "stringVal": [
                    AUTHOR
                  ]
                }
              }
            }
          ]
        },
        "citationMetadata": {
          "listVal": [
            {
              "structVal": {
                "citations": {}
              }
            }
          ]
        },
        "safetyAttributes": {
          "listVal": [
            {
              "structVal": {
                "categories": {},
                "blocked": {
                  "boolVal": [
                    false
                  ]
                },
                "scores": {}
              }
            }
          ]
        }
      }
    }
  ]
}

Vertex AI SDK for Python

For information about installing the Vertex AI SDK for Python, see Install the Vertex AI SDK for Python.

Text 

  import vertexai
  from vertexai.language_models import TextGenerationModel

  def streaming_prediction(
      project_id: str,
      location: str,
  ) -> str:
      """Streaming Text Example with a Large Language Model"""

  vertexai.init(project=project_id, location=location)

  text_generation_model = TextGenerationModel.from_pretrained("text-bison")
  parameters = {
      "temperature": temperature,  # Temperature controls the degree of randomness in token selection.
      "max_output_tokens": 256,  # Token limit determines the maximum amount of text output.
      "top_p": 0.8,  # Tokens are selected from most probable to least until the sum of their probabilities equals the top_p value.
      "top_k": 40,  # A top_k of 1 means the selected token is the most probable among all tokens.
  }

  responses = text_generation_model.predict_streaming(prompt="Give me ten interview questions for the role of program manager.", **parameters)
  for response in responses:
      `print(response)`

Chat 

import vertexai
from vertexai.language_models import ChatModel, InputOutputTextPair

def streaming_prediction(
    project_id: str,
    location: str,
) -> str:
    """Streaming Chat Example with a Large Language Model"""

    vertexai.init(project=project_id, location=location)

    chat_model = ChatModel.from_pretrained("chat-bison")

    parameters = {
        "temperature": 0.8,  # Temperature controls the degree of randomness in token selection.
        "max_output_tokens": 256,  # Token limit determines the maximum amount of text output.
        "top_p": 0.95,  # Tokens are selected from most probable to least until the sum of their probabilities equals the top_p value.
        "top_k": 40,  # A top_k of 1 means the selected token is the most probable among all tokens.
    }

    chat = chat_model.start_chat(
        context="My name is Miles. You are an astronomer, knowledgeable about the solar system.",
        examples=[
            InputOutputTextPair(
                input_text="How many moons does Mars have?",
                output_text="The planet Mars has two moons, Phobos and Deimos.",
            ),
        ],
    )

    responses = chat.send_message_streaming(
        message="How many planets are there in the solar system?", **parameters)
    for response in responses:
        `print(response)`

Code 

import vertexai
from vertexai.language_models import CodeGenerationModel

def streaming_prediction(
    project_id: str,
    location: str,
) -> str:
    """Streaming Chat Example with a Large Language Model"""

    vertexai.init(project=project_id, location=location)

    code_model = CodeGenerationModel.from_pretrained("code-bison")
    parameters = {
        "temperature": 0.8,  # Temperature controls the degree of randomness in token selection.
        "max_output_tokens": 256,  # Token limit determines the maximum amount of text output.
    }

    responses = code_model.predict_streaming(
        prefix="Write a function that checks if a year is a leap year.", **parameters)
    for response in responses:
        `print(response)`

Code chat 

import vertexai
from vertexai.language_models import CodeChatModel

def streaming_prediction(
    project_id: str,
    location: str,
) -> str:
    """Streaming Chat Example with a Large Language Model"""

    vertexai.init(project=project_id, location=location)

    codechat_model = CodeChatModel.from_pretrained("codechat-bison")
    parameters = {
        "temperature": 0.8,  # Temperature controls the degree of randomness in token selection.
        "max_output_tokens": 1024,  # Token limit determines the maximum amount of text output.
    }
    codechat = codechat_model.start_chat()

    responses = codechat.send_message_streaming(
        message="Please help write a function to calculate the min of two numbers", **parameters)
    for response in responses:
        `print(response)`

Available client libraries

You can use one of the following client libraries to stream responses:

  • Python
  • Node.js
  • Java
  • Go
  • C#

To view sample code requests and responses using the REST API, see Examples using the REST API.

To view sample code requests and responses using the Vertex AI SDK for Python, see Examples using Vertex AI SDK for Python.

Responsible AI

Responsible Artificial Intelligence (RAI) filters scan the streaming output as the model generates it. If a violation is detected, the filters block the offending output tokens, and return an output with a blocked flag under safetyAttributes, which terminates the stream.

What's next