GLM 5V Turbo Is Now Available in Puter.js

Puter.js now supports GLM-5V Turbo, Z.AI's native multimodal coding model designed to bridge visual perception and code generation in a single architecture.

What is GLM-5V Turbo?

GLM-5V Turbo processes images, video, and text natively and is optimized for agentic workflows — turning design mockups, screenshots, and UI layouts into runnable code. Built on the GLM-5 architecture (744B total parameters, 40B active in a Mixture of Experts design), it combines strong vision capabilities with coding performance. Key highlights include:

• Design-to-Code Leader: Scores 94.8 on the Design2Code benchmark (vs. Claude Opus 4.6's 77.3), making it the top model for converting visual designs into code
• GUI Agent Performance: Leads on GUI agent benchmarks like AndroidWorld and WebVoyager, outperforming Claude Opus 4.5 on BrowseComp for agentic browsing
• Native Multimodal Input: Processes images, video, and text natively without separate preprocessing pipelines
• 200K Context Window: Handle large documents, long videos, and complex multi-step workflows
• RL-Trained Across 30+ Task Types: Maintains strong text-only coding performance alongside its vision strengths

Examples

Design mockup to code

puter.ai.chat(
  "Convert this UI design into responsive HTML and CSS",
  "https://example.com/mockup.png",
  { model: 'z-ai/glm-5v-turbo' }
);

Image analysis

puter.ai.chat(
  "What do you see in this image?",
  "https://assets.puter.site/doge.jpeg",
  { model: 'z-ai/glm-5v-turbo' }
);

Video analysis

puter.ai.chat(
  "Describe what happens in this video and identify the key moments",
  "https://assets.puter.site/puppy.mp4",
  { model: 'z-ai/glm-5v-turbo' }
);

Screenshot to bug report

puter.ai.chat(
  "Analyze this screenshot of our app. Identify any UI issues, accessibility problems, or layout bugs and generate a detailed bug report",
  "https://example.com/app-screenshot.png",
  { model: 'z-ai/glm-5v-turbo' }
);

Document parsing

puter.ai.chat(
  "Extract all the data from this invoice image and return it as structured JSON",
  "https://example.com/invoice.png",
  { model: 'z-ai/glm-5v-turbo' }
);

Get Started Now

Just add one library to your project:

// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

Or add one script tag to your HTML:

<script src="https://js.puter.com/v2/"></script>

No API keys needed. Start building with GLM-5V Turbo immediately.

Learn more:

• Free, Unlimited Z.AI GLM API
• Puter.js AI Documentation