Gemini (native format) — File Analysis

Summary: Gemini native-format requests support mixed file and text inputs via the contents array. This page provides the minimal required fields, example requests, and response examples for calling Gemini native endpoints on this platform (camelCase parameter names).

---

Overview

• Method: POST
• Path: /v1beta/models/{model}:{action} (example: /v1beta/models/gemini-2.5-flash:generateContent)
• Content-Type: application/json

Authentication

• Supported authentication methods:
  • Authorization: Bearer sk-xxx (recommended)
  • x-goog-api-key: <api-key> (optional)

---

Path parameters

Parameter	Required	Description
model	yes	Model name, e.g. gemini-2.5-flash
action	yes	Action name, e.g. generateContent (non-stream) or streamGenerateContent?alt=sse (stream)

---

Request body (structure and notes)

• Top-level JSON field: contents (required, array). Each element is a content object, typically containing role and parts.

Common request fields:

Field	Type	Required	Description
contents	array	yes	Array of request contents; each item contains role and parts
contents[].role	string	yes	Role, typically user / system / assistant
contents[].parts	array	yes	Array of parts; items are objects that may include text for textual content

Note: Gemini native format supports additional optional fields (thinking controls, extraBody, tools, etc.). The platform standardizes on camelCase parameter names (e.g. toolConfig instead of tool_config).

---

Examples

Simple text example (curl)

curl -X POST "https://api.gpt.ge/v1beta/models/gemini-2.5-flash:generateContent" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-xxxx" \
  -d '{
    "contents": [
      { "role": "user", "parts": [{ "text": "Hello, please give a brief self-introduction" }] }
    ]
  }'

File analysis example (curl, file using base64 data URI)

curl -X POST "https://api.gpt.ge/v1beta/models/gemini-2.5-pro:generateContent" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-xxxx" \
  -d '{
    "contents": [
      {
        "role": "user",
        "parts": [
          { "text": "Please summarize the key points in the attached API document" },
          { "file": { "filename": "api-doc.pdf", "fileData": "data:application/pdf;base64,JVBERi0xLjQKJ..." } }
        ]
      }
    ]
  }'

JavaScript example

await fetch('https://api.gpt.ge/v1beta/models/gemini-2.5-flash:generateContent', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json', 'Authorization': 'Bearer sk-xxxx' },
  body: JSON.stringify({ contents: [{ role: 'user', parts: [{ text: 'Hello, please give a brief self-introduction' }] }] })
}).then(r => r.json()).then(console.log)

Python example (requests)

import requests

payload = {
  'contents': [
    { 'role': 'user', 'parts': [ { 'text': 'Please summarize this document' } ] }
  ]
}

resp = requests.post('https://api.gpt.ge/v1beta/models/gemini-2.5-flash:generateContent', headers={'Content-Type':'application/json','Authorization':'Bearer sk-xxxx'}, json=payload)
print(resp.json())

---

Response example (200)

{
  "id": "resp-1234",
  "object": "model.response",
  "created": 1725014307,
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "This is a brief summary of the document: ..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": { "prompt_tokens": 50, "completion_tokens": 120, "total_tokens": 170 }
}

---

Server example: https://api.gpt.ge

For more advanced parameters (thinking level, extraBody, tools, etc.), see Gemini's official docs or other example pages in the project.