Gemini API

The Gemini API provides access to Google's Gemini models. The endpoint URL includes the model name in the path, and authentication can be passed via URL parameter or header.

Endpoint Info

Item	Value
URL	/v1beta/models/{model}:generateContent
Method	POST
Auth	URL parameter key or Bearer Token header
Content-Type	application/json

URL Format

Replace {model} with the model name:

• /v1beta/models/gemini-2.5-flash:generateContent
• /v1beta/models/gemini-2.5-pro:generateContent

For streaming, use :streamGenerateContent instead.

Request Parameters

Parameter	Type	Required	Description
contents	array	Yes	Message array with role and parts
system_instruction	object	No	System instruction (single part object)
generation_config	object	No	Generation settings (temperature, max_tokens, etc.)
safety_settings	array	No	Safety filter settings

contents Format

{
  "contents": [
    {
      "role": "user",
      "parts": [
        {"text": "Hello, Gemini!"}
      ]
    },
    {
      "role": "model",
      "parts": [
        {"text": "Hello! How can I assist you?"}
      ]
    },
    {
      "role": "user",
      "parts": [
        {"text": "Tell me about quantum computing"}
      ]
    }
  ]
}

generation_config Fields

Field	Type	Description
temperature	number	Sampling temperature, 0–2
max_output_tokens	integer	Maximum output tokens
top_p	number	Nucleus sampling
top_k	integer	Top-k sampling
candidate_count	integer	Number of candidates to generate

cURL Examples

URL Parameter Authentication

curl "https://api.quickapi.store/v1beta/models/gemini-2.5-flash:generateContent?key=YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [
      {
        "role": "user",
        "parts": [{"text": "Hello, Gemini!"}}
      }
    ],
    "generation_config": {
      "temperature": 0.7,
      "max_output_tokens": 1000
    }
  }'

Header Authentication

curl https://api.quickapi.store/v1beta/models/gemini-2.5-flash:generateContent \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "contents": [
      {
        "role": "user",
        "parts": [{"text": "Hello, Gemini!"}}
      }
    ],
    "generation_config": {
      "temperature": 0.7,
      "max_output_tokens": 1000
    }
  }'

Streaming (streamGenerateContent)

curl "https://api.quickapi.store/v1beta/models/gemini-2.5-flash:streamGenerateContent?key=YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [
      {
        "role": "user",
        "parts": [{"text": "Write a short poem"}}
      }
    ],
    "generation_config": {
      "max_output_tokens": 500
    }
  }'

With System Instruction

curl https://api.quickapi.store/v1beta/models/gemini-2.5-pro:generateContent \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "contents": [
      {
        "role": "user",
        "parts": [{"text": "Explain this code: def fib(n): return n if n <= 1 else fib(n-1) + fib(n-2)"}}
      }
    ],
    "system_instruction": {
      "parts": [{"text": "You are an expert Python programmer. Explain code clearly."}]
    },
    "generation_config": {
      "temperature": 0.3,
      "max_output_tokens": 2000
    }
  }'

💡 Model Name in Path

Unlike OpenAI's API where the model is specified in the request body, Gemini requires the model name in the URL path. Make sure to use the correct model identifier.

Response Example

Normal Response

{
  "candidates": [
    {
      "content": {
        "role": "model",
        "parts": [
          {
            "text": "Hello! I'm Gemini, Google's AI model. How can I help you today?"
          }
        ]
      },
      "finishReason": "STOP",
      "index": 0,
      "safetyRatings": [
        {
          "category": "HARM_CATEGORY_SEXUALLY_EXPLICIT",
          "probability": "NEGLIGIBLE"
        },
        {
          "category": "HARM_CATEGORY_HATE_SPEECH",
          "probability": "NEGLIGIBLE"
        },
        {
          "category": "HARM_CATEGORY_HARASSMENT",
          "probability": "NEGLIGIBLE"
        },
        {
          "category": "HARM_CATEGORY_DANGEROUS_CONTENT",
          "probability": "NEGLIGIBLE"
        }
      ]
    }
  ],
  "usageMetadata": {
    "promptTokenCount": 6,
    "candidatesTokenCount": 20,
    "totalTokenCount": 26
  }
}

Streaming Response

[
  {
    "candidates": [
      {
        "content": {
          "role": "model",
          "parts": [{"text": "Roses are red"}]
        },
        "finishReason": "STOP"
      }
    ]
  },
  {
    "candidates": [
      {
        "content": {
          "role": "model",
          "parts": [{"text": ",
violets are blue"}]
        }
      }
    ]
  },
  {
    "candidates": [
      {
        "content": {
          "role": "model",
          "parts": [{"text": ",
AI is great"}]
        }
      }
    ]
  },
  {
    "candidates": [
      {
        "content": {
          "role": "model",
          "parts": [{"text": ",
and so are you!"}]
        },
        "finishReason": "STOP"
      }
    ],
    "usageMetadata": {
      "promptTokenCount": 5,
      "candidatesTokenCount": 18,
      "totalTokenCount": 23
    }
  }
]

Python Example

import google.generativeai as genai

genai.configure(
    api_key="YOUR_API_KEY",
    client_options={"api_endpoint": "https://api.quickapi.store"}
)

model = genai.GenerativeModel("gemini-2.5-flash")

# Normal request
response = model.generate_content("Hello, Gemini!")
print(response.text)

# With generation config
response = model.generate_content(
    "Explain quantum computing",
    generation_config=genai.types.GenerationConfig(
        temperature=0.7,
        max_output_tokens=2000,
        top_p=0.9,
        top_k=40
    )
)
print(response.text)

# Streaming request
for chunk in model.generate_content(
    "Write a short poem",
    stream=True
):
    print(chunk.text, end="")

# With system instruction
model_with_system = genai.GenerativeModel(
    "gemini-2.5-pro",
    system_instruction="You are an expert Python programmer."
)
response = model_with_system.generate_content("Explain decorators in Python")
print(response.text)

⚠️ API Version

This endpoint uses the v1beta API version. Some features may change before reaching stable v1. Check Google's official documentation for the latest updates.