17h ago

Google releases Gemini Omni video generation demonstrations

296415520096.1K

——0——

Google released early demonstrations of Gemini Omni, a new multimodal video generation model. The samples feature a professor writing and explaining trigonometric identities on a chalkboard, including equations such as Sin²(x) + Cos²(x) = 1, while preserving step-by-step text accuracy and scene coherence. Additional clips reference a lamp scene from SeeDance 2. The demonstrations are accessible at gemini.google.com/share/7d5dc678c80a.

Original post

Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)#582@TEORTAXESTEX

Google took SeeDance 2 personally https://twitter.com/chetaslua/status/2053824398503678108

7:06 AM · May 11, 2026

Cluster engagement

94 snapshots

AI 1000 · 2 actions

QUOTET(#582@TEORTAXESTEX Google took SeeDance 2 personally https://twitter.com/chetaslua/status/2053824398503678108
REPOSTKD#959@CSPROFKGD @TOMLIKESROBOTSVery interesting. A new multimodal video model from Google apparently called Gemini Omni. (https://gemini.google.com/share/7d5dc678c80a) For interest, the one with the lamp showing is SeeDance 2. Gemini does seem to do better at Maths. Prompt: A professor writes out a mathematical proof for trigonometric identities on a traditional chalkboard, explaining the step he is currently on in the equation.
Very interesting. A new multimodal video model from Google apparently called Gemini Omni. (https://gemini.google.com/share/7d5dc678c80a) For interest, the one with the lamp showing is SeeDance 2. Gemini does seem to do better at Maths. Prompt: A professor writes out a mathematical proof for trigonometric identities on a traditional chalkboard, explaining the step he is currently on in the equation.