SVG T2I
SVG-T2I is a text-to-image diffusion model that generates directly in Visual Foundation Model feature space (e.g. DINOv3) instead of pixels or VAE latents, achieving high-fidelity images and competitive GenEval and DPG-Bench scores.
- Type
- Image
- Weights
- Closed
- Downloads
- 21
- Likes
- 35
- Pipeline
- text to image
- License
- apache-2.0
Model hub
KlingTeam/SVG-T2I
- Repo
- KlingTeam/SVG-T2I
- Library
- pytorch
- Access
- Public
Technical tags and languages
Papers
2
More from Kuaishou Technology
11
-
41k DL
KAT Coder
Multimodal
KAT-Coder is Kwaipilot’s strongest code intelligence model, designed for advanced software engineering and ag…
-
KAT Coder Pro V2
Multimodal
KAT-Coder-Pro V2 is a hosted Kwaipilot coding model listed on OpenRouter with a 256,000-token context window.…
-
KAT Dev 32B
Multimodal
KAT-Dev-32B is Kwaipilot’s open-source 32B software engineering model built for real-world coding tasks. It i…
-
Kling 1.5
Video
Kling 1.5 is Kuaishou’s upgraded text-to-video system. It generates sharper, more coherent footage with bette…
-
Kling 1.6
Video
Kling 1.6 is Kuaishou’s incremental upgrade to its text-to-video system, delivering cleaner motion, better id…
-
Kling 2.0
Video
Kling 2.0 is Kuaishou’s next-gen text-to-video model. It produces sharper, longer, and more stable clips with…
-
Kling 2.1
Video
Kling 2.1 is Kuaishou’s latest text-to-video update, pushing 2.0’s quality further with crisper detail, stron…
-
Kling 2.1 Master
Video
Maximum fidelity profile for demanding shots, tuned for strong physics, complex choreography, and long sequen…
About Kuaishou Technology
1
Chinese short-video & live-streaming platform (Kwai) with e-commerce and ads.