GLM-4.1V-Thinking-Flash API

Name: GLM-4.1V-Thinking-Flash
Brand: Zhipu AI
Availability: InStock

A free vision-language model that thinks before it answers: visual chain-of-thought reasoning over images, documents, charts, and screenshots, with a 128K context window.

FREE

Input / 1M tokens

FREE

Output / 1M tokens

128K

Context window

Image

+ text input

Why use GLM-4.1V-Thinking-Flash on NovAI?

Free forever - zero cost for image and text tokens alike
Visual chain-of-thought - reasons step-by-step about image content before answering, boosting accuracy on complex visuals
Strong OCR & document parsing - reads dense text, tables, and forms in screenshots and scans
Chart & diagram understanding - extracts data and logic from visualizations
128K context - analyze multi-page documents and many images in one call
Standard OpenAI vision format - send image_url content parts, no proprietary SDK

Not listed on OpenRouter - free vision-reasoning models of this class are exclusive to the Chinese providers. NovAI gives you direct official access with USD billing.

Best use cases

Screenshot-to-code: turn UI mockups into HTML/CSS/React
Document intelligence: invoices, contracts, forms
Chart question-answering and data extraction
Automated UI testing and visual QA
Educational tools that explain diagrams step by step

Quick start

cURL (image + text)

curl https://aiapi-pro.com/v1/chat/completions \
  -H "Authorization: Bearer $NOVAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-4.1v-thinking-flash",
    "messages": [{
      "role": "user",
      "content": [
        {"type": "text", "text": "Extract every line item from this invoice"},
        {"type": "image_url", "image_url": {"url": "https://example.com/invoice.png"}}
      ]
    }]
  }'

Python (OpenAI SDK)

from openai import OpenAI
client = OpenAI(base_url="https://aiapi-pro.com/v1", api_key="YOUR_NOVAI_API_KEY")
resp = client.chat.completions.create(
    model="glm-4.1v-thinking-flash",
    messages=[{
        "role": "user",
        "content": [
            {"type": "text", "text": "What does this chart tell us?"},
            {"type": "image_url", "image_url": {"url": "https://example.com/chart.png"}},
        ],
    }],
)
print(resp.choices[0].message.content)

Works with any OpenAI-compatible SDK - just set base_url to https://aiapi-pro.com/v1 and use your NovAI API key. Full reference in the API docs.