Video models

Google's third Omni video generation and editing model, raising one generation from 10 to 15 seconds and cutting the 720p price by about 30% against Omni 1.1 Flash.

Released

Nov 5, 2026

$ / sec

$0.070

Max length

15 s

HeyGen's first general-purpose video model, built on MiniMax H3 and post-trained by HeyGen. It generates a whole scene with picture and sound in one call, from a text prompt, a first-frame image or reference files.

Released

Sep 30, 2026

$ / sec

—

Max length

30 s

Utopai Studios' video generation model, used inside its PAI production-intelligence platform for film and television. Filmmakers select it to generate and compare takes for a shot.

Released

Sep 30, 2026

$ / sec

—

Max length

—

Google's updated Omni video generation and editing model, adding scene extension, first-and-last-frame control and 1080p/4K upscaled output to Gemini Omni Flash.

Released

Aug 27, 2026

$ / sec

$0.10

Max length

10 s

Speed-oriented variant of MiniMax H3, post-trained by fal on top of the open H3 weights and released jointly by MiniMax and fal; MiniMax's API exposes it as a separate model ID.

Released

Aug 27, 2026

$ / sec

$0.080

Max length

15 s

Alibaba's closed, API-only Wan video model: up to 30-second clips at up to 1080p with audio, from text, images, audio, video and documents or web pages.

Released

Aug 24, 2026

$ / sec

$0.20

Max length

30 s

MiniMax's omni-modal video model: it reads text, images, video and audio as one context and generates video with native stereo audio, up to 2K and 15 seconds. The 768p base model's weights are on Hugging Face.

Released

Jul 31, 2026

$ / sec

$0.080

Max length

15 s

ByteDance Seed's audio-video generation model, sold as Dreamina Seedance 2.5: up to 30-second clips, up to 50 omni-reference assets, and timestamp-based editing and extension.

Released

Jul 31, 2026

$ / sec

—

Max length

30 s

Black Forest Labs' multimodal foundation model, trained jointly on images, video and audio. This entry covers its video generation (FLUX 3 Video), which makes clips of up to 20 seconds with native audio from text, images, keyframes or an input clip.

Released

Jul 23, 2026

$ / sec

$0.17

Max length

20 s

Meta Superintelligence Labs' video generation model with native audio, built on the same pretraining base as Muse Image and shown as an early preview.

Released

Jul 7, 2026

$ / sec

—

Max length

—

xAI's video generation model, generally available on the Imagine API, generating video with sound effects, ambience and lip-synced dialogue in a single pass.

Released

Jun 16, 2026

$ / sec

$0.080

Max length

15 s

The first model in Google's Omni family: a multimodal model that generates video from mixed text, image, audio and video inputs and edits it conversationally over multiple turns.

Released

May 20, 2026

$ / sec

$0.10

Max length

10 s

Video-and-audio generation model from Alibaba's ATH (Alibaba Token Hub) business unit: multi-shot clips up to 15 seconds at 1080p with synchronized audio, including lip-synced dialogue.

Released

Apr 28, 2026

$ / sec

$0.24

Max length

15 s

Skywork AI's joint video-audio model (dual-stream MMDiT) that generates, inpaints and edits video from text, images, video clips, masks and audio references, at up to 1080p, 32 FPS and 15 seconds.

Released

Feb 25, 2026

$ / sec

$0.35

Max length

15 s

ByteDance Seed's joint audio-video generation model, sold as Dreamina Seedance 2.0: takes text, image, audio and video inputs for multimodal reference, editing and extension, at 4-15 seconds.

Released

Feb 12, 2026

$ / sec

—

Max length

15 s

Google DeepMind's Veo video model with native audio, adding reference images, first/last-frame transitions and scene extension; offered with Fast and Lite variants.

Released

Oct 15, 2025

$ / sec

$0.40

Max length

8 s

Sapiens AI's video generation model, served through the Agnes AI platform's OpenAI Videos-compatible API. It supports text-to-video, first/last-frame keyframe control and multimodal reference generation.

Released

Undated

$ / sec

$0.025

Max length

12 s

Speed-oriented post-train of FLUX 3, built by fal Research with Black Forest Labs on the FLUX 3 base. It generates text- and image-to-video with native audio, a 10-second clip in about 12 seconds.

Released

Undated

$ / sec

$0.17

Max length

20 s