whatmodel.sh / image / CogView4-6B
CogView4-6B
↗ share
+ contribute
CogView4-6B is an open-weights image model from Z.ai, released March 2025. Its published size is 6B parameters. Licence: apache-2.0.
Catalogue identity notes
Zhipu/Z.ai CogView4 text-to-image model, 6B DiT (zai-org/CogView4-6B, pinned 63a52b7, Apache-2.0; the only CogView4 release). MMMG Table 2 'CogView-4' (1024 px, DM) and the HPSv3 README benchmark 'CogView4'. Comparison-only.
Model information sources
Release date ↗ · Weights licence ↗
6B as named by the maker (CogView4-6B DiT); GLM-4-9B text encoder separate.
Score within the LMArena text-to-image comparison set only.
$ how it compares
The image models ranked either side of CogView4-6B, out of 197 with a score, then more from Z.ai.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 140 |  Playground v2.5 | 41±2 | local | ||
| 141 |  Stable Diffusion 3.5 Medium | 41±3 | api $0.035/image +local | ||
| 142 | CogView4-6B | 41±6 | comparison only | ||
| 143 |  Midjourney V7 | 40±3 | comparison only | ||
| 144 |  UniWorld-V1 | 40±2 | comparison only | ||
| more from Z.ai | |||||
| 102 | GLM-Image | 50±2 | api $0.015/image +local | ||
$ what it's good at
text to image▁▂▇▇▆▅█▇▂▃▁▁ #110 of 157 * · 87% of its score · 2 tests
Beats Latent Diffusion text2img-large (LDM-KL-8-G, LAION-400M), VQ-Diffusion (ITHQ), OmniGen2 here, though they rank higher overall. Trails Show-o2 1.5B-HQ here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| MMMG knowledge-image generation (MMMG-Score, average over education levels) | 13.1% ✎ | ▃··▃·█▃▃▅··▃ beats 3/9 | 77% |
| HPSv3 image generation benchmark (Overall HPSv3 score) | 9.6100 ✎ | ▃·····▃··▃·█ beats 2/5 | 10% |
$ where to run it
No verified API or download recorded yet · add one
Weights and documentation: huggingface.co/zai-org ↗
$ how the score was worked out · MMMG: A Massive, Multidiscipl… 77%, HPSv3 repository README, imag… 10%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 MMMG: A Massive, Multidisciplinary, Multi-Tier Generation Benchmark for Text-to-Image Reasoning (arXiv:2506.10963v2) | 1 | 77% |
 HPSv3 repository README, image generation benchmark table (MizzenAI/HPSv3) | 1 | 10% |
| prior · pull toward the middle | 13% | |
| weighted average | = 40.6 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json