Getting Started with the Gemini API in Vertex AI with cURL / REST API

Once Google Cloud project is set up and the Vertext AI API is enabled, you can use the Vertex AI REST API with cURL commands to interact with the Gemini 2.0 Flash model.

 

 

 

- Text generation

The `generateContent` method can handle a wide variety of cases including below.

- multi-turn chat

- multimodal input

 

Below example, you send a text prompt and request the model response in text.

%%bash

curl -X POST \
  -H "Authorization: Bearer $(gcloud auth print-access-token)" \
  -H "Content-Type: application/json" \
  https://${API_ENDPOINT}:generateContent \
  -d '{
    "contents": {
      "role": "USER",
      "parts": { "text": "Why is the sky blue?" },
    },
    "generation_config": {
      "response_modalities": "TEXT",
     },
  }' 2>/dev/null >response.json

jq -r ".candidates[].content.parts[].text" response.json

 

 

- Streaming text generation

The Gemini API provides a streaming response mechanism. It's like you instantly get the answer from chat GPT when you ask a question in the prompt. You don't need to wait for the complete response. 

%%bash

curl -X POST \
  -H "Authorization: Bearer $(gcloud auth print-access-token)" \
  -H "Content-Type: application/json" \
  https://${API_ENDPOINT}:streamGenerateContent \
 \
  -d '{
    "contents": {
      "role": "USER",
      "parts": { "text": "Why is the sky blue?" }
    }
  }' 2>/dev/null >response.json

jq -r ".[] | .candidates[] | .content.parts[].text" response.json

 

 

- Chat

 

The Gemini API supports multi-turn conversations where user can get back-and-forth interactions with a model.

Specify the `role` field only if the content represents a turn in a conversation.

%%bash

curl -X POST \
  -H "Authorization: Bearer $(gcloud auth print-access-token)" \
  -H "Content-Type: application/json" \
  https://${API_ENDPOINT}:generateContent \
  -d '{
    "contents": [
      {
        "role": "user",
        "parts": [
          { "text": "Hello" }
        ]
      },
      {
        "role": "model",
        "parts": [
          { "text": "Hello! I am glad you could both make it." }
        ]
      },
      {
        "role": "user",
        "parts": [
          { "text": "So what is the first order of business?" }
        ]
      }
    ]
  }' 2>/dev/null >response.json

jq -r ".candidates[].content.parts[].text" response.json

 

 

- Function Calling

When you ask a question or command to a model, function calling creates a description of a function in ther code, then pass that description to a language model.

%%bash

curl -X POST \
  -H "Authorization: Bearer $(gcloud auth print-access-token)" \
  -H "Content-Type: application/json" \
  https://${API_ENDPOINT}:generateContent \
  -d '{
  "contents": {
    "role": "user",
    "parts": {
      "text": "Which theaters in Mountain View show Barbie movie?"
    }
  },
  "tools": [
    {
      "function_declarations": [
        {
          "name": "find_movies",
          "description": "find movie titles currently playing in theaters based on any description, genre, title words, etc.",
          "parameters": {
            "type": "object",
            "properties": {
              "location": {
                "type": "string",
                "description": "The city and state, e.g. San Francisco, CA or a zip code e.g. 95616"
              },
              "description": {
                "type": "string",
                "description": "Any kind of description including category or genre, title words, attributes, etc."
              }
            },
            "required": [
              "description"
            ]
          }
        },
        {
          "name": "find_theaters",
          "description": "find theaters based on location and optionally movie title which are is currently playing in theaters",
          "parameters": {
            "type": "object",
            "properties": {
              "location": {
                "type": "string",
                "description": "The city and state, e.g. San Francisco, CA or a zip code e.g. 95616"
              },
              "movie": {
                "type": "string",
                "description": "Any movie title"
              }
            },
            "required": [
              "location"
            ]
          }
        },
        {
          "name": "get_showtimes",
          "description": "Find the start times for movies playing in a specific theater",
          "parameters": {
            "type": "object",
            "properties": {
              "location": {
                "type": "string",
                "description": "The city and state, e.g. San Francisco, CA or a zip code e.g. 95616"
              },
              "movie": {
                "type": "string",
                "description": "Any movie title"
              },
              "theater": {
                "type": "string",
                "description": "Name of theater"
              },
              "date": {
                "type": "string",
                "description": "Date for requested showtime"
              }
            },
            "required": [
              "location",
              "movie",
              "theater",
              "date"
            ]
          }
        }
      ]
    }
  ]
}' 2>/dev/null >response.json

jq -r ".candidates[].content.parts[].functionCall" response.json

 

 

- Multimodal Input

Gemin supports adding image and video in chat prompts.

 

 

Ref

Gemini for Google Cloud Learning Path > Course > Develop GenAI Apps with Gemini and Streamlit