10 proprietary AI video models from 5 labs. Input is priced at $1.50 per 1M tokens. The largest context window is 1M tokens (Gemini Omni 1.1 Flash). Newest check: 9 Nov 2026.
94 of 197 models were checked against the vendor's own page in the last 30 days.
| Model | Lab | Context | In / out per 1M | Reads and writes | Status | Checked |
|---|---|---|---|---|---|---|
| Genie 4Genie 4 is Google DeepMind's interactive world model, announced in October 2026, with native audio, spoken dialogue and scenes steered by text prompts. | -- | -- | text, image → video, audio | Preview | 29 Oct | |
| Kling 4.0Kling 4.0 is Kuaishou's AI video model for long single-take clips with multi-keyframe control, Omni Reference and stereo audio, announced in September 2026. | Kuaishou Technology | -- | -- | text, image, video, audio → video, audio | Available | 5 Nov |
| Gemini Omni 1.1 FlashGemini Omni 1.1 Flash (Google, GA 2026) generates and edits AI video through chat, adding native synchronized audio and multi-pass scene extension. | 1M | $1.50 / $17.50 | text, image, video → video, audio | Available | 7 Oct | |
| Wan 3.0Tongyi Lab's public-beta video generator (since August 6, 2026) makes native single-shot clips up to half a minute long from text, image, audio, video or documents. | Alibaba Cloud | -- | -- | text, image, audio, video, pdf → video | Beta | 26 Oct |
| Seedance 2.5Seedance 2.5 is ByteDance's 2026 video generation model, producing one continuous multi-reference clip per pass with synchronized audio, launched inside Jimeng AI and Doubao Pro. | ByteDance | -- | -- | text, image, video, audio → video, audio | Available | 14 Sep |
| Gemini OmniSuperseded by Gemini Omni 1.2 Flash on 5 November 2026; this original Omni Flash from May 2026 is still what Google's AI Plus, Pro and Ultra plans use for video in the Gemini app and Google Flow. | -- | -- | text, image, audio, video → video, audio, image, text | Preview | 9 Nov | |
| HappyHorse 1.0Built by Alibaba's ATH Business Unit, HappyHorse 1.0 generates synchronized video and audio together from a text, image or reference-image prompt. | Alibaba Cloud | -- | -- | text, image, video → video, audio | Available | 26 Sep |
| Veo 3.1Google DeepMind's Veo 3.1 generates 4, 6 or 8-second video clips with native synchronized dialogue and audio, rendering up to 4K. | -- | -- | text, image → video, audio | Preview | 26 Sep | |
| FLUX 3FLUX 3 is a video, image, and audio model from Black Forest Labs, launched July 2026, built on a shared Self-Flow architecture. | Black Forest Labs | -- | -- | text, image, video → video, audio | Preview | 18 Oct |
| Kling 3.0Kling 3.0 is Kuaishou Technology's proprietary AI video model, built for native 4K output with synchronized multilingual audio in one generation pass. | Kuaishou Technology | -- | -- | text, image, video → video, audio | Available | 30 Aug |
A person checks every record against the vendor's own page. Rankings cannot be bought. How Hokai checks records. Not sure what you need? Describe the job to Smart Match and get three picks with reasons.