Z.AI offers a variety of models and agents to meet the needs of different scenarios. Choosing the right model can help you complete tasks more efficiently.
Featured Models
GLM-4.6
Our New Flagship Model
Designed for Agent Applications
GLM-4.5V
Breakthrough in Open-Source VLM
Support Switching Thinking Modes
CogVideoX-3
More Stable and Clearer
New Start & End Frame Generation
Models, Agents and Tools
To help you find the best fit for your use case, we’ve created a table outlining the core features and strengths of each model in the Z.AI family.Text Models
Our model matrix includes text models with built-in reasoning capabilities, as well as vision-language models (VLMs) that extend the same reasoning power to multimodal understanding.| Model | Strength | Language | Context | Resourse |
|---|---|---|---|---|
| GLM-4.6 | Highest Performance Strong Coding More Versatile | English & Chinese | 200K | Guide API Reference |
| GLM-4.5 | Better Performance Strong Reasoning More Versatile | English & Chinese | 128K | Guide API Reference |
| GLM-4.5V(vlm) | Multimodal Flexible Reasoning State-of-the-art in its scale | English & Chinese | 64K | Guide API Reference |
| GLM-4.5-X | Good Performance Strong Reasoning Ultra-Fast Response | English & Chinese | 128K | Guide API Reference |
| GLM-4.5-Air | Cost-Effective Lightweight High Performance | English & Chinese | 128K | Guide API Reference |
| GLM-4.5-AirX | Lightweight High Performance Ultra-Fast Response | English & Chinese | 128K | Guide API Reference |
| GLM-4-32B-0414-128K | High intelligence at unmatched cost-efficiency | English & Chinese | 128K | Guide API Reference |
| GLM-4.5-Flash | Lightweight High Performance | English & Chinese | 128K | Guide API Reference |
Built-in Tools
A suite of built-in tools designed to streamline workflows and boost productivity.| Tool | Capability |
|---|---|
| Web Search | - Provide real-time, concise, direct answers - Accurately parse complex HTML and converts it into clean Markdown or JSON |
Image Generation Models
Image Generation Models learn from massive image data to automatically generate high-quality images from text.| Model | Strength | Language | Resolution | Resourse |
|---|---|---|---|---|
| CogView-4 | - High-quality image generation - Diverse styles - Rich in detail | English & Chinese | multiple resolutions | Guide API Reference |
Video Generation Models
Video Generation Models turn text, images, or clips into dynamic video content, accelerating creativity for film, virtual avatars, animation, and marketing.| Model | Strength | Language | Resolution | Resourse |
|---|---|---|---|---|
| CogVideoX-3 | Significant improvements in image quality, stability, and physical realism simulation | English & Chinese | multiple resolutions | Guide API Reference |
| ViduQ1 | Theatrical quality with seamless temporal flow | English & Chinese | 1080P | Guide API Reference |
| Vidu2 | Fast delivery with smart style preservation | English & Chinese | 720P | Guide API Reference |
Agents
A set of ready-made agents empower users to create and communicate effortlessly.| Tool | Capability | Resource |
|---|---|---|
| GLM Slide/Poster Agent(beta) | Combine content generation with professional design | Guide |
| General-Purpose Translation | Support 40+ languages, flexible strategies, and terminology customization | Guide |
| Popular Special Effects Video Templates | Special effects video templates like French_Kiss, BodyShake, and Sexy_Me | Guide |