# Is Kandinsky S2V 1.0 Pro SOTA?

Pending benchmarks. Announced Oct 13, 2026; no independent results yet.

As of Nov 22, 2026, Kandinsky S2V 1.0 Pro is pending benchmarks on isitsota: announced Oct 13, 2026, with no independent results on the boards we track yet.

The 21B size of Kandinsky Lab's Kandinsky S2V 1.0 speech-to-video family: it animates a person from a single reference image, a speech track and a text prompt, with lip motion, expression, pose and gesture following the speech. It does not generate audio. Output is 25 fps, SD or HD, with long speech generated in 5-second chunks. Initialized from Kandinsky 5.0 Video. Code and weights under MIT.

Canonical: https://isitsota.com/m/kandinsky-s2v-1-0-pro
Last updated: 2026-11-04
Tracking since: 2026-11-04

## Announced facts

| Fact | Value |
| --- | --- |
| Lab | Kandinsky |
| Announced | Oct 13, 2026 |
| API | Not yet |
| Open weights | Published |
| Parameters | 21B |
| Task | Speech-to-video (audio-driven avatar animation) from one reference image, a speech waveform and a text prompt; audio is a condition only |
| Output | 25 fps video, SD or HD at the reference frame's aspect ratio; long speech in 5-second chunks |
| Architecture | 3D Diffusion Transformer with flow matching in a frozen Hunyuan video VAE latent; Qwen2.5-Omni text and audio front end; three-scale FramePack visual memory |
| Training data | 1.17M single-speaker clips, captions in English and Russian |
| Checkpoints | Base, pi-Flow distilled (16 steps instead of 50) and Diffusers checkpoints on Hugging Face; demo Space |
| License | MIT (code and weights) |
| Related models | Separate from Kandinsky S2V 1.0 Lite and from Kandinsky 6.0 Video |

Tags marked unsure are not stated on the lab's own pages.

## What would let us call it

Verdicts use only independent boards we already track. Two consecutive snapshots are required before any crown.

- [x] Publicly announced with a source
- [x] Usable by the public (API or weights)
- [ ] Listed on an independent board we track
- [ ] Enough results for a rank in a category
- [ ] Holds across two snapshots (crowns only)

## What the lab claims

Self-reported. Shown for context; not used in any verdict.

- Lab claim: in a side-by-side human evaluation, Kandinsky S2V 1.0 Pro performs comparably to or better than the strongest open-source baseline and two proprietary commercial models on average, and is preferred for lip sync.
- Lab claim: the distilled Pro model runs in 16 steps without classifier-free guidance, a 5.4x end-to-end speedup over the base model.

## Sources

- PAPER: [Kandinsky S2V 1.0: Audio-Driven Avatar Video Generation (arXiv)](https://arxiv.org/abs/2610.19290) (2026-10-13)
- REPO: [kandinskylab/Kandinsky-S2V-1.0-Pro model card (Hugging Face)](https://huggingface.co/kandinskylab/Kandinsky-S2V-1.0-Pro)
- REPO: [kandinskylab/kandinsky-s2v (GitHub)](https://github.com/kandinskylab/kandinsky-s2v)
- LAB: [Kandinsky S2V 1.0 project page](https://kandinskyavatar.github.io/)

Tracking continues until an independent board lists it.

## Related

- Lab: [Kandinsky](https://isitsota.com/labs/kandinsky.md)
- All pending releases: https://isitsota.com/releases.md
- JSON: https://api.isitsota.com/v1/models/kandinsky-s2v-1-0-pro
