Skip to main content

Sora alternatives

Updated 2026-11-19

Short answer

OpenAI discontinued the Sora web and app experiences on April 26, 2026 and the Sora API on September 24, 2026. The main replacements are other text-to-video models that also make sound: Google's Veo 3.1 and Gemini Omni, Kuaishou's Kling 4.0, ByteDance's Seedance, MiniMax's Hailuo, Alibaba's Wan and HappyHorse, and xAI's Grok Imagine. On MakeMidas they share one credit balance: no subscription to each maker, and you pay only for what you make, at the prices in the table.

The alternatives

a hummingbird hovers at a red flower in a sunny garden, wings beating, birdsong
Veo 3.1By Google · From 96 credits per 4s clip

Google DeepMind's video model. Veo 3.1 Quality renders up to 4K, always with sound; Fast and Lite are the 720p and 1080p versions, with sound you can switch off.

macro shot in a dollhouse kitchen: a tiny hamster chef in a white toque flips a miniature pancake from a tiny pan, the pancake spins once in slow motion and lands on a small stack, warm morning window light, shallow depth of field, static camera. Sound: butter sizzling, a soft plop, a tiny bell ding
Kling AIBy Kuaishou · From 96 credits per 3s clip

Kuaishou's model. Kling 4.0 makes 3 to 15 second clips at up to 1080p with sound on every clip, from text or a first and last frame.

a barista pours latte art into a cup, close-up, steam rising, warm cafe light
Seedance 2.0By ByteDance · From 60 credits per 4s clip

ByteDance's model. It takes reference images as well as first and last frames, makes 4 to 15 second clips with optional sound, and has lower-cost Fast, Mini and Flash versions.

extreme macro of a small frog made of translucent green jelly sitting on a wet leaf, a single raindrop lands on its head and its whole body wobbles, then it blinks, soft overcast light, glossy refractions, static camera. Sound: gentle rain and one wet plop
Gemini Omni FlashBy Google · From 9 credits per 3s clip

Google's Gemini Omni Flash, from 360p up to 4K with sound on every clip, and 3 to 10 seconds long.

a corgi in a small white spacesuit floating weightless inside a sunlit space station module, it paddles its paws and catches a floating strawberry in its mouth, Earth glowing blue through the round window behind, cinematic, slow gentle camera drift
Hailuo AIBy MiniMax · From 128 credits per 4s clip

MiniMax's model. Hailuo 3 renders up to 2K with sound on every clip and runs up to 15 seconds; the Max versions are silent and go up to 1080p.

a street musician plays violin under a streetlight at night, light rain
Wan 3.0By Alibaba · From 32 credits per 4s clip

Alibaba's Wan 3.0 and 3.0 Prime: 4 to 12 second clips at up to 1080p, with optional sound and first and last frames.

a horse knitted from chunky cream wool gallops across rolling hills made of patchwork quilts, yarn mane bouncing, felt clouds in a pastel sky, handmade stop-motion look, camera tracking alongside at the same speed
HappyHorseBy Alibaba · From 96 credits per 3s clip

Alibaba's HappyHorse: 3 to 15 second clips at up to 1080p, from text, a first frame or reference images.

a giant dragon made of folded book pages rises out of the waterfall of pages, unfurls its paper wings and glides between the floating library islands, loose pages swirling in its wake, tiny readers on the rope bridges look up in awe, golden sunset light, sweeping cinematic camera
Grok ImagineBy xAI · From 64 credits per 4s clip

xAI's Grok Imagine: version 1.5 makes 4 to 15 second clips with sound, from text or reference images.

Compared at a glance

What each model offers on MakeMidas today, read from the same settings and prices as the generator.

ModelMakerUp toLengthSoundPrice on MakeMidas
Veo 3.1Google4K4–8 sYesFrom 36 credits per 4s clip
Kling AIKuaishou1080p3–15 sYesFrom 54 credits per 3s clip
Seedance 2.0ByteDance1080p4–15 sYesFrom 32 credits per 4s clip
Gemini Omni FlashGoogle4K3–10 sYesFrom 9 credits per 3s clip
Hailuo AIMiniMax2K4–15 sYesFrom 42 credits per 6s clip
Wan 3.0Alibaba1080p4–12 sYesFrom 32 credits per 4s clip
HappyHorseAlibaba1080p3–15 sYesFrom 96 credits per 3s clip
Grok ImaginexAI1080p4–15 sYesFrom 24 credits per 4s clip

What happened to Sora

OpenAI's help article gives April 26, 2026 for the end of the Sora web and app experiences and September 24, 2026 for the Sora API. It also explains how to export your old Sora videos before the data is deleted.

How to pick a replacement

For 4K, look at Veo 3.1 Quality and Gemini Omni Flash. For clips longer than 10 seconds, Kling 4.0, Seedance, Hailuo, HappyHorse and Grok Imagine 1.5 go up to 15 seconds here. To keep a character or product the same across shots, use a model that takes reference images, such as Seedance or HappyHorse. To start from a picture, nearly every model here takes a first frame, and most also a last frame.

Price depends on the model, resolution and length; the table above shows where each starts, and the AI video cost calculator prices any clip.

From ChatGPT, too

If you made Sora videos from ChatGPT, you can connect MakeMidas to ChatGPT and ask for a video there with any of these models; see Can ChatGPT make videos?.

Questions

Is Sora still available?

No. OpenAI's help center says the Sora web and app experiences were discontinued on April 26, 2026, and the Sora API on September 24, 2026.

Which Sora alternative makes 4K video?

On MakeMidas, Veo 3.1 Quality and Gemini Omni Flash render up to 4K.

Which Sora alternatives make video with sound?

Every family here has a version that makes sound. Some add it to every clip, such as Kling 4.0, Veo 3.1 Quality and Gemini Omni Flash; others, such as Seedance, Wan and Veo 3.1 Fast and Lite, have a sound switch.

Do I need an account with each company?

No. One MakeMidas account and one credit balance cover every model here, in the website, in ChatGPT and Claude, and through the API.

Sources

From the maker

  1. What to know about the Sora discontinuation · OpenAI Help Center
  2. Veo · Google DeepMind
  3. Kling 4.0 AI video generation · Kuaishou