CogView Models by Zhipu AI

Zhipu AIProprietary
Zhipu AI releases · 15 in the last 12 monthsChangelog →
3 models2022

Last refreshed 2026-07-11. Next refresh: weekly.

Details

ResearcherZhipu AI
LicenseProprietary
Commercial useRestricted
Models3
Released2022

About

CogView is Zhipu AI's family of text-to-image models. CogView3 uses cascaded relay diffusion, while CogView-3Plus moved to a Diffusion Transformer design. CogView4 is a 6B open-weight model that replaces the English-only T5 text encoder with Zhipu's bilingual GLM-4 encoder, so it understands Chinese and English prompts and can write Chinese characters inside images; Zhipu calls it its first open-source text-to-image model able to do so. It generates images at many resolutions, from 512 to 2048 pixels per side. On Zhipu's BigModel platform, the family is offered as the CogView-4 API and the free CogView-3-Flash model.

Decision facts

Best fit
image generationvisioncoding
Capability starting point
CogView 2
Lowest tracked input
Not tracked
Closest related family
AutoGLM

Current Variants

Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.

3 in view
CogView 2Current

Use when the workload needs 6B parameters.

2022-106B parameters

Use when the workload needs image generation and vision.

Unknown releaseimage generationvision
CogView-4Current

Use when the workload needs image generation and vision.

Unknown releaseimage generationvision

Release Timeline

2 release groups
2022-10
1 current
CogView 2
6B parameters
Current
Unknown release
2 current
CogView-3-Flash
image generationvision
Current
CogView-4
image generationvision
Current

Specifications(3 models)

CogView model specifications comparison
ModelReleasedParameters
CogView 22022-106B
CogView-3-Flash——
CogView-4——

Available From(2 providers)