POST
/
v1beta
/
models
/
gemini-2.5-pro:streamGenerateContent
curl -X POST "https://api.example.com/v1beta/models/gemini-2.5-pro:streamGenerateContent?key=YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [
      {
        "parts": [{"text": "1+2+3+4+5+....+999?"}],
        "role": "user"
      }
    ],
    "generationConfig": {
      "temperature": 1,
      "topP": 1,
      "thinkingConfig": {
        "includeThoughts": true,
        "thinkingBudget": 26240
      }
    }
  }'

{
  "candidates": [
    {
      "content": {
        "parts": [
          {"thought": "Let me calculate the sum..."},
          {"text": "The sum is 499500."}
        ],
        "role": "model"
      },
      "finishReason": "STOP"
    }
  ]
}
Official documentation: https://ai.google.dev/gemini-api/docs/text-generation
Streaming text generation with thinking process using Google Gemini 2.5 Pro model.

​
Request Parameters

​
key
string
required
API key.
​
contents
array
required
Conversation content array.
​
systemInstruction
object
System instruction configuration.
​
safetySettings
array
Safety settings array. Configurable categories include:
  • HARM_CATEGORY_HATE_SPEECH
  • HARM_CATEGORY_SEXUALLY_EXPLICIT
  • HARM_CATEGORY_HARASSMENT
  • HARM_CATEGORY_DANGEROUS_CONTENT
  • HARM_CATEGORY_CIVIC_INTEGRITY
​
generationConfig
object
Generation configuration.
  • temperature (number): Sampling temperature
  • topP (number): Nucleus sampling parameter
  • thinkingConfig (object): Thinking configuration
    • includeThoughts (boolean): Whether to include thinking process
    • thinkingBudget (integer): Thinking budget (maximum 26240)
curl -X POST "https://api.example.com/v1beta/models/gemini-2.5-pro:streamGenerateContent?key=YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [
      {
        "parts": [{"text": "1+2+3+4+5+....+999?"}],
        "role": "user"
      }
    ],
    "generationConfig": {
      "temperature": 1,
      "topP": 1,
      "thinkingConfig": {
        "includeThoughts": true,
        "thinkingBudget": 26240
      }
    }
  }'

{
  "candidates": [
    {
      "content": {
        "parts": [
          {"thought": "Let me calculate the sum..."},
          {"text": "The sum is 499500."}
        ],
        "role": "model"
      },
      "finishReason": "STOP"
    }
  ]
}