Skip to main content

​
Gemini OpenAI-Compatible Format

You can call Gemini models through the standard OpenAI Chat Completions API format without learning the Gemini native API. 
POST /v1/chat/completions

​
Basic Conversation

curl https://crazyrouter.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "gemini-2.5-flash",
    "messages": [
      {"role": "system", "content": "You are a knowledgeable assistant."},
      {"role": "user", "content": "Explain quantum entanglement"}
    ],
    "temperature": 0.7,
    "max_tokens": 1024
  }'

​
Vision Understanding

Python
response = client.chat.completions.create(
    model="gemini-2.5-flash",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "Describe this image"},
                {
                    "type": "image_url",
                    "image_url": {"url": "https://example.com/photo.jpg"}
                }
            ]
        }
    ],
    max_tokens=1024
)

​
File Understanding

Send PDFs and other files through the OpenAI-compatible format:
Python
import base64

with open("document.pdf", "rb") as f:
    pdf_base64 = base64.b64encode(f.read()).decode("utf-8")

response = client.chat.completions.create(
    model="gemini-2.5-flash",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "Summarize this document"},
                {
                    "type": "image_url",
                    "image_url": {
                        "url": f"data:application/pdf;base64,{pdf_base64}"
                    }
                }
            ]
        }
    ],
    max_tokens=2048
)

​
Image Generation

Use Gemini image generation models through the OpenAI-compatible format:
Python
response = client.chat.completions.create(
    model="gemini-2-5-flash-image",
    messages=[
        {
            "role": "user",
            "content": "Draw a cute cat playing in a garden"
        }
    ],
    max_tokens=4096
)

# The response may contain Base64 images
print(response.choices[0].message.content)

​
Streaming Output

Python
stream = client.chat.completions.create(
    model="gemini-2.5-flash",
    messages=[
        {"role": "user", "content": "Implement a simple HTTP server in Go"}
    ],
    max_tokens=2048,
    stream=True
)

for chunk in stream:
    if chunk.choices[0].delta.content is not None:
        print(chunk.choices[0].delta.content, end="")

​
Available Gemini Models

ModelDescription
gemini-2.5-proGemini 2.5 Pro, strongest capability
gemini-2.5-flashGemini 2.5 Flash, fast response
gemini-2.5-flash-thinkingFlash with thinking capability
gemini-2-5-flash-imageFlash with image generation support
gemini-3-pro-image-previewGemini 3 Pro image preview
When using Gemini through the OpenAI-compatible format, Crazyrouter automatically converts requests to Gemini native format and converts responses back to OpenAI format. Most features work normally, but certain Gemini-specific features (such as code execution, Google Search) require using the native format.