Open-weight image generation models
18
Models
13 organisations
100%
Open weights
18 open, 0 closed
0
Frontier
By Epoch's flag
27B
Largest traceable
Qwen Image
2018–2025
Span
10 with a traceable size
Organisation
Records
18| Model | Organisation | Released | Parameters | Weights |
|---|---|---|---|---|
| Qwen Image | Alibaba | 2025-08 | 27B | Open |
| Infinity | ByteDance | 2024-12 | 2B | Open |
| CTM (CIFAR-10) | Stanford University | 2023-10 | — | Open |
| Stable Diffusion XL (SDXL) | Stability AI | 2023-07 | 3B * | Open |
| DiT-XL/2 | New York University (NYU) | 2023-03 | 675M | Open |
| DDPM-IP (CelebA) | Utrecht University | 2023-01 | 295M * | Open |
| Discriminator Guidance | Korea Advanced Institute of Science and Technology (KAIST) | 2022-11 | — | Open |
| Diffusion-GAN | UT Austin | 2022-06 | — | Open |
| Stable Diffusion (LDM-KL-8-G) | Runway | 2022-04 | 1B | Open |
| LDM-1.45B | Heidelberg University | 2021-12 | 1B | Open |
| Projected GAN | Heidelberg University | 2021-11 | — | Open |
| StyleGAN3-R | NVIDIA | 2021-06 | 2M | Open |
| StyleGAN3-T | NVIDIA | 2021-06 | 2M | Open |
| ADM | OpenAI | 2021-05 | 559M | Open |
| CogView | Tsinghua University | 2021-05 | 4B * | Open |
| VQGAN + CLIP | Heidelberg University | 2020-12 | — | Open |
| StyleGAN2 | NVIDIA | 2019-12 | 30M | Open |
| StyleGAN | NVIDIA | 2018-12 | 26M | Open |
An asterisk marks a parameter figure the source does not consider traceable to the people who built the model. That judgement is the source’s and it has false negatives: Kimi K3 is marked speculative at 2.8T while Moonshot’s own model card states the figure. Why that distinction carries the whole exhibit.
Counts are of models Epoch AI records as notable, not of every model released. The most recent month is always incomplete, because a model enters the dataset when it is reviewed rather than when it ships.
Epoch AI, 'Data on AI Models'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/data/ai-models-documentation' Used under CC BY 4.0.