Image-to-Video models for Physical AI: autonomous driving, robotics, general physics.
AI & ML interests
Gen AI, AIGC, Video Generation, Image Generation
Recent Activity
View all activity
KVAE 2.0 is a family of image and video tokenizers with a time compression ratio of 4 and spacial compression ratio of 8 and 16
-
kandinskylab/Kandinsky-5.0-T2I-Lite-sft-Diffusers
Text-to-Image • 6B • Updated • 386 • 17 -
kandinskylab/Kandinsky-5.0-I2I-Lite-sft-Diffusers
Image-to-Image • 6B • Updated • 131 • 8 -
kandinskylab/Kandinsky-5.0-I2I-Lite-pretrain-Diffusers
Image-to-Image • 6B • Updated • 21 • 3 -
kandinskylab/Kandinsky-5.0-T2I-Lite-pretrain-Diffusers
Text-to-Image • 6B • Updated • 16 • 1
KVAE 1.0 tokenizers are for images (KVAE-2D-1.0) and video (KVAE-3D-1.0) are distributed under MIT license (commercial use is possible).
Kandinsky 5.0 Video Pro is a 19B model that generates high-quality HD videos from English and Russian prompts with controllable camera motion.
KVAE-Audio is a continuous full-band audio waveform autoencoder
Kandinsky 5.0 Video Pro is a 19B model that generates high-quality HD videos from English and Russian prompts with controllable camera motion.
-
kandinskylab/Kandinsky-5.0-I2V-Pro-sft-5s-Diffusers
19B • Updated • 49 • 33 -
kandinskylab/Kandinsky-5.0-T2V-Pro-sft-5s-Diffusers
19B • Updated • 34 • 8 -
kandinskylab/Kandinsky-5.0-I2V-Pro-distilled-5s-Diffusers
19B • Updated • 31 • 12 -
kandinskylab/Kandinsky-5.0-T2V-Pro-distilled-5s-Diffusers
19B • Updated • 12 • 6
-
kandinskylab/Kandinsky-5.0-I2V-Pro-LoRa-Microwave-right
Image-to-Video • Updated • 1 -
kandinskylab/Kandinsky-5.0-I2V-Pro-LoRa-Microwave-left
Image-to-Video • Updated -
kandinskylab/Kandinsky-5.0-T2V-Pro-LoRa-Microwave-right
Text-to-Video • Updated -
kandinskylab/Kandinsky-5.0-T2V-Pro-LoRa-Microwave-left
Text-to-Video • Updated
Kandinsky 5.0 Video Lite is a lightweight 2B model that generates up to 10-second SD videos from English and Russian prompts with high visual quality.
-
kandinskylab/Kandinsky-5.0-T2V-Lite-distilled16steps-10s-Diffusers
2B • Updated • 20 -
kandinskylab/Kandinsky-5.0-T2V-Lite-distilled16steps-5s-Diffusers
2B • Updated • 62 • 1 -
kandinskylab/Kandinsky-5.0-T2V-Lite-nocfg-10s-Diffusers
2B • Updated • 3 -
kandinskylab/Kandinsky-5.0-T2V-Lite-nocfg-5s-Diffusers
2B • Updated • 19 • 1
Kandinsky 5.0 Image Lite is a 6B DiT-based model that generates and edits HD images from English and Russian text prompts with high visual quality.
Kandinsky 5.0 Video Lite is a lightweight 2B model that generates up to 10-second SD videos from English and Russian prompts with high visual quality.
Image-to-Video models for Physical AI: autonomous driving, robotics, general physics.
KVAE-Audio is a continuous full-band audio waveform autoencoder
KVAE 2.0 is a family of image and video tokenizers with a time compression ratio of 4 and spacial compression ratio of 8 and 16
Kandinsky 5.0 Video Pro is a 19B model that generates high-quality HD videos from English and Russian prompts with controllable camera motion.
-
kandinskylab/Kandinsky-5.0-I2V-Pro-sft-5s-Diffusers
19B • Updated • 49 • 33 -
kandinskylab/Kandinsky-5.0-T2V-Pro-sft-5s-Diffusers
19B • Updated • 34 • 8 -
kandinskylab/Kandinsky-5.0-I2V-Pro-distilled-5s-Diffusers
19B • Updated • 31 • 12 -
kandinskylab/Kandinsky-5.0-T2V-Pro-distilled-5s-Diffusers
19B • Updated • 12 • 6
-
kandinskylab/Kandinsky-5.0-T2I-Lite-sft-Diffusers
Text-to-Image • 6B • Updated • 386 • 17 -
kandinskylab/Kandinsky-5.0-I2I-Lite-sft-Diffusers
Image-to-Image • 6B • Updated • 131 • 8 -
kandinskylab/Kandinsky-5.0-I2I-Lite-pretrain-Diffusers
Image-to-Image • 6B • Updated • 21 • 3 -
kandinskylab/Kandinsky-5.0-T2I-Lite-pretrain-Diffusers
Text-to-Image • 6B • Updated • 16 • 1
-
kandinskylab/Kandinsky-5.0-I2V-Pro-LoRa-Microwave-right
Image-to-Video • Updated • 1 -
kandinskylab/Kandinsky-5.0-I2V-Pro-LoRa-Microwave-left
Image-to-Video • Updated -
kandinskylab/Kandinsky-5.0-T2V-Pro-LoRa-Microwave-right
Text-to-Video • Updated -
kandinskylab/Kandinsky-5.0-T2V-Pro-LoRa-Microwave-left
Text-to-Video • Updated
Kandinsky 5.0 Video Lite is a lightweight 2B model that generates up to 10-second SD videos from English and Russian prompts with high visual quality.
-
kandinskylab/Kandinsky-5.0-T2V-Lite-distilled16steps-10s-Diffusers
2B • Updated • 20 -
kandinskylab/Kandinsky-5.0-T2V-Lite-distilled16steps-5s-Diffusers
2B • Updated • 62 • 1 -
kandinskylab/Kandinsky-5.0-T2V-Lite-nocfg-10s-Diffusers
2B • Updated • 3 -
kandinskylab/Kandinsky-5.0-T2V-Lite-nocfg-5s-Diffusers
2B • Updated • 19 • 1
KVAE 1.0 tokenizers are for images (KVAE-2D-1.0) and video (KVAE-3D-1.0) are distributed under MIT license (commercial use is possible).
Kandinsky 5.0 Image Lite is a 6B DiT-based model that generates and edits HD images from English and Russian text prompts with high visual quality.
Kandinsky 5.0 Video Pro is a 19B model that generates high-quality HD videos from English and Russian prompts with controllable camera motion.
Kandinsky 5.0 Video Lite is a lightweight 2B model that generates up to 10-second SD videos from English and Russian prompts with high visual quality.