Introducing CogVideoX-3Z.AIReleased July 15, 2025

CogVideoX-3

Advanced video generation model featuring start and end frame synthesis, improved physical realism simulation, and high visual stability.

text-to-videoProprietary API$0.20 / video (Keyframe Synthesis)Context: 4.096K (4,096 tokens)

Technical Specifications

Architecture Type
Diffusion Video Transformer with 3D VAE
Total Parameters
Frontier Scale Video DiT
Context Window
4.096K (4,096 tokens)
Max Output Tokens
1.024K (1,024 tokens)
Knowledge Cutoff
June 2025
Supported Modalities
video, text, vision
License & Access
Z.AI Terms of Service

Benchmark Evaluations

Motion Continuity Score
92.1
Physical Simulation Score
87.5

Deep Architectural Overview

CogVideoX-3 represents Z.AI’s third-generation video architecture. It incorporates 3D Variational Autoencoders and temporal attention to generate fluid, physically consistent video clips from text or keyframes at $0.20 per video.

Strengths & Considerations

Core Strengths
  • Start and end frame interpolation for controlled storytelling
  • Substantial improvements in physical consistency and object stability
  • High resolution video generation
Known Limitations
  • Audio synchronization requires separate audio pairing

Token & API Pricing

Input Tokens (1M)Contact
Output Tokens (1M)Contact
Pricing is verified directly against Z.AI's developer documentation and API rate sheets.

Similar & Alternative Models

Explore other frontier models from Z.AI and comparable reasoning engines.

Browse all models
Z.AI

Ultra-low latency version of GLM-5.3-Flash engineered for high-concurrency real-time IDE code completion and interactive agents.

200K ctx$0.37 / 1M tok (Ultra-Low Latency MoE)
Z.AI

The first native multimodal model in the GLM-5 series: 320B parameters (18B active) combining linear and sparse attention for low-cost visual coding.

200K ctx$0.15 / 1M tok (320B MoE Visual Coding)
Z.AI

Z.AI’s premier flagship model, delivering a 50% performance gain on Code Bench and matching Claude Mythos 5 in cybersecurity and vulnerability discovery.

1M ctx$1.40 / 1M tok (Flagship Coding & Cyber)
Z.AI

Flagship model built for project-scale context, supporting truly usable 1M-token context with 128k output.

1M ctx$1.40 / 1M tok (1M Lossless Context)