Project AURIONAI Use-Cases Research & Innovation
Sber
image generation

Kandinsky 2.2

Open text-to-image model from Sber AI that maps text to an image embedding with a prior and decodes it with latent diffusion.

SberSberRussia
Portrait image generated by Kandinsky 2.2.
Levelmodel
Classimage generation
Parameters1.25B
Contextnot in the notes
Released9 Jun 2023
Opennessopen weights permissive
Runs onnot in the notes
Statusactive

R&D Topics

Text-to-image foundation models (diffusion and flow transformers)produced by

Identity

Developer
Sber
Family
not in the notes
Released
9 Jun 2023 sourcesregistry dateunverified
Weights published
–
Record checked
12 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
diffusion Image prior plus latent diffusion decoder
Total parameters
1.25B sources
Active parameters
–
Checkpoint parameters
–
Note
Parameter count from the Hugging Face safetensors total.
Context length
not in the notes
Input
text
Output
image
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
open weights permissive
Licence
Apache License 2.0 sources
Access routes
open weights download, API
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
not in the notes
Footprint
not in the notes
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
not stated
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Benchmark Results

The notes record no benchmark result for this model.

Lineage

The notes record no lineage for this model.

Industry Use Cases

No industry use case links this model.

Sources

  1. huggingface.co/kandinsky-community/kandinsky-2-2-decoder https://huggingface.co/kandinsky-community/kandinsky-2-2-decoder
  2. arxiv.org/abs/2310.03502 https://arxiv.org/abs/2310.03502
  3. replicate.com/ai-forever/kandinsky-2.2 https://replicate.com/ai-forever/kandinsky-2.2

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Kandinsky 2.2", https://aurion.semai.id/models/sber-kandinsky-2-2/, records of October 2026.