Multimodal (Vision)

AI Foundation Services provides vision models that can analyze images alongside text. Use the same Chat Completions API with image content.

Prerequisites

• An API key (get one here)
• OpenAI SDK or HTTP client installed (Quickstart)
• A vision-capable model (see Available Models)

What you'll learn:

• How to analyze images from URLs
• How to send local images via base64 encoding
• Which models support vision capabilities

---

Analyze an Image from URL

curl

curl -X POST "$OPENAI_BASE_URL/chat/completions" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-2.5-flash",
    "messages": [
      {
        "role": "user",
        "content": [
          {"type": "text", "text": "What is in this image?"},
          {"type": "image_url", "image_url": {"url": "https://images.unsplash.com/photo-1546069901-ba9599a7e63c?w=400"}}
        ]
      }
    ],
    "max_tokens": 1024
  }'

Python

from openai import OpenAI

client = OpenAI()

response = client.chat.completions.create(
    model="gemini-2.5-flash",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "What's in this image?"},
                {
                    "type": "image_url",
                    "image_url": {
                        "url": "https://images.unsplash.com/photo-1546069901-ba9599a7e63c?w=400"
                    },
                },
            ],
        }
    ],
    max_tokens=1024,
)

print(response.choices[0].message.content)

Node.js

import OpenAI from "openai";

const client = new OpenAI();

const response = await client.chat.completions.create({
  model: "gemini-2.5-flash",
  messages: [
    {
      role: "user",
      content: [
        { type: "text", text: "What's in this image?" },
        {
          type: "image_url",
          image_url: {
            url: "https://images.unsplash.com/photo-1546069901-ba9599a7e63c?w=400",
          },
        },
      ],
    },
  ],
  max_tokens: 1024,
});

console.log(response.choices[0].message.content);

---

Analyze a Local Image (Base64)

You can also pass a local image as a base64-encoded string:

import base64
from openai import OpenAI

client = OpenAI()

def encode_image(image_path):
    with open(image_path, "rb") as f:
        return base64.b64encode(f.read()).decode("utf-8")

base64_image = encode_image("/path/to/your/image.jpg")

response = client.chat.completions.create(
    model="Qwen3-VL-30B-A3B-Instruct-FP8",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "What's in this image?"},
                {
                    "type": "image_url",
                    "image_url": {"url": f"data:image/jpeg;base64,{base64_image}"},
                },
            ],
        }
    ],
    max_tokens=1000,
)

print(response.choices[0].message.content)

---

Available Vision Models

Model	Provider	Capabilities
Qwen3-VL-30B-A3B-Instruct-FP8	T-Cloud (Germany)	Image understanding, OCR
gemini-2.5-flash	Google Cloud	Image + video understanding
gpt-4.1	Azure	Image understanding

Check Available Models for the latest list.

---

Next Steps

• Visual RAG — Index and retrieve from documents with text + image understanding
• Function Calling — Connect models to external tools
• Streaming — Stream responses for better UX