Kandinsky Enters Physical AI with Open-Source WM 1.0 Release
kabachuha · reddit · 2026-08-04
Kandinsky lab has released the Kandinsky WM 1.0 model family, officially entering the 'Physical AI' domain. The project is open-source on GitHub. The community is optimistic about the release, speculating that the promised Kandinsky 6.0 Audio-Video model might be coming soon.
More from Multimodal
- RTX 5090 Test: Minimax H3 Takes 50 Mins for 1.5MP Image — SourceTraining7959 · 2026-08-04
- Goodbye Short Cuts: Testing Seedance 2.5 for Seamless Long-Take AI Video — Uncanny_Harry · 2026-08-04
- Hands-on Comparison: MMH3 Outperforms Google Omni Flash — Ok-Act-9620 · 2026-08-04
- Krea2 Tested with Hiroshi Nagai Inspired Style LoRA — dataiwarrior · 2026-08-04
- MiniMax-H3 I2V Test: ref2va Matches fl2va with Single Image Input — spacemidget75 · 2026-08-04
- Running MiniMax-H3 on RTX 3060: Renders 5s Video in 4 Minutes — Robert_Brown_7425 · 2026-08-04