The AI Museum
AI Analytics1,072 recordsRecounted 23 September 2026

Image generation models

Weightsopenclosedunknownfrontier only
Year20262025202420232022202120202019201820172016
1 filter active, showing 47 of 1,072 recordsClear all

47

Models

24 organisations

40%

Open weights

18 open, 27 closed

0

Frontier

By Epoch's flag

27B

Largest traceable

Qwen Image

2014–2026

Span

15 with a traceable size

Releases

by year
2014201520162017201820192020202120222023202420252026

Organisation

Country

Domain

Records

47
Image generation models, newest first. Each row links to the source Epoch AI records for it.
ModelOrganisationReleasedParametersWeights
ChatGPT Images 2.5OpenAI2026-09—Closed
GPT-Image-2.5 FlareOpenAI2026-09—Closed
GPT-Image-2.5 SunburstOpenAI2026-09—Closed
MAI-Image-2.5Microsoft2026-06—Closed
GPT Image 2OpenAI2026-04—Closed
Seedance 2.0ByteDance2026-02—Closed
Gemini 3 Pro Image (Nano Banana Pro)Google DeepMind2025-11—Closed
Gemini 2.5 Flash Image (Nano Banana)Google2025-08—Closed
Qwen ImageAlibaba2025-0827BOpen
Imagen 4Google2025-05—Closed
gpt-image-1OpenAI2025-04—Closed
InfinityByteDance2024-122BOpen
Stable Diffusion 3Stability AI2024-028BClosed
CTM (CIFAR-10)Stanford University2023-10—Open
DALL·E 3OpenAI2023-10—Closed
DiT-XL/2 + CADSETH Zurich2023-10675M *Closed
Stable Diffusion XL (SDXL)Stability AI2023-073B *Open
DiT-XL/2New York University (NYU)2023-03675MOpen
FireflyAdobe2023-03—Closed
DDPM-IP (CelebA)Utrecht University2023-01295M *Open
AR-LDMAlibaba2022-112BClosed
Discriminator GuidanceKorea Advanced Institute of Science and Technology (KAIST)2022-11—Open
DiT-XL/2 + Discriminator GuidanceKorea Advanced Institute of Science and Technology (KAIST)2022-11—Closed
eDiff-INVIDIA2022-119B *Closed
Diffusion-GANUT Austin2022-06—Open
PartiGoogle Research2022-0620BClosed
ImagenGoogle Brain2022-058B *Closed
DALL·E 2OpenAI2022-044B *Closed
Stable Diffusion (LDM-KL-8-G)Runway2022-041BOpen
XMC-GANGoogle Research2022-04—Closed
Make-A-SceneMeta AI2022-034B *Closed
MaskGIT (ImageNet)Google Research2022-02227MNot recorded
Midjourney V1Midjourney2022-02—Closed
LDM-1.45BHeidelberg University2021-121BOpen
Projected GANHeidelberg University2021-11—Open
StyleGAN3-RNVIDIA2021-062MOpen
StyleGAN3-TNVIDIA2021-062MOpen
ADMOpenAI2021-05559MOpen
CogViewTsinghua University2021-054B *Open
DALL-EOpenAI2021-0112B *Closed
VQGAN + CLIPHeidelberg University2020-12—Open
StyleGAN2NVIDIA2019-1230MOpen
PG-SWGANETH Zurich2019-06—Closed
SPN (ImageNet 128)Google Brain2018-12250MClosed
StyleGANNVIDIA2018-1226MOpen
WGAN-GPCourant Institute of Mathematical Sciences2017-03—Closed
GANsUniversity of Montreal / Université de Montréal2014-06—Not recorded

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.

How this museum is made