Sora 2 vs Veo 3.1
OpenAI's Sora 2 and Google's Veo 3.1 were the two flagship video models of late 2025. One is being retired on 24 September 2026; the other has three tiers and 4K. We ran both through the same nine clips before the shutdown.
Anvisha Pai, Co-founder & CEO, Voyager
Which to use
Veo 3.1, and it is not close on the clips or the calendar. Veo typeset the title card that Sora set in wobbling thin type, moved the speaking woman where Sora's barely moved, and threw a spill with real droplets. Sora 2 held its own on the turntable and the market and produced the most charming self-portrait on the site, but every Sora clip is four seconds, 720p only, and the API disappears on 24 September 2026. If price is the objection, Veo 3.1 Fast gives most of the full model's quality at $0.15 a second, only half again Sora's rate.
What others found
Other people who ran both models, quoted and linked. Where our clips agree or disagree, we say so.
After seven rounds, Veo 3.1 takes it with consistent prompt adherence and audio complexity. While Sora 2 often looked better and felt more atmospheric, it frequently ignored difficult audio-visual instructions.
Ryan Morrison, Tom's Guide, seven audio prompts through both models, .
Matches our set: Veo followed the title card and the spoken line where Sora looked handsome and did less.
Sora creates believable environments; Veo follows the script.
Ryan Morrison, Tom's Guide, .
Veo 3.1 still has the edge on raw visual realism for single shots. Runway Gen-4's world consistency across separate generations is more mature. Sora 2 handles complex camera movements well.
Alisaqqt, Hacker News comment, after testing across several hosts, .
Side by side, measured
| Sora 2 | Veo 3.1 | |
|---|---|---|
| Vendor | OpenAI | Google DeepMind |
| Released | 2025-09-30 | 2025-10-15 |
| Price | $0.10 per second at 720p, sora-2 via OpenAI API | $0.40 per second with audio, 720p or 1080p via Gemini API |
| Seconds to make one clip | 67 to 200 | 67 to 108 |
| 9 clips cost | $3.60 | $21.60 |
| Length | 4, 8, 12, 16 or 20 seconds on fal's endpoint; our five-second fixture rounds down to four | 4, 6 or 8 seconds only; our five-second fixture rounds to six |
Seconds and cost are from our own runs of the same 9 prompts through each model.
The same 9 prompts, Sora 2 against Veo 3.1
One try each, nothing re-rolled or retouched, generated by us. Sora 2 is on the left, Veo 3.1 on the right. Every prompt has its own page with the whole roster.
Product turntable
Every model on this prompt →Both hold the dripper through the turn; Sora's is the tidier of the two.
›What we asked for
A matte black ceramic pour-over coffee dripper rotating slowly on a white turntable, soft studio light, seamless white background, product video, no text.
Person speaking
Every model on this prompt →Sora's speaker turns her head and smiles and little else; Veo invented a selfie framing, arm outstretched, with a full delivery.
›What we asked for
A woman in her 40s with short grey hair stands in a bright kitchen, looks at the camera and says 'Good morning, the coffee is ready', natural window light, handheld feel, she smiles at the end.
Animated title card
Every model on this prompt →Veo set both lines on a card and held them; Sora's thin type wobbled and lost letters by the last frame.
›What we asked for
A title card reading 'MODA SUMMER SESSIONS' animates in over slow-motion footage of a festival crowd at dusk, then a smaller line 'Friday 12 June' fades in beneath it. Bold condensed type, tangerine on navy.
Camera move
Every model on this prompt →›What we asked for
Drone shot pushing slowly forward over a coastal town at sunrise, terracotta roofs, fishing boats in the harbour, long shadows, cinematic and steady.
Physics
Every model on this prompt →Veo's spill spreads with droplets; Sora's is competent and backlit.
›What we asked for
A glass of water tips over on a wooden table and spills in slow motion, water and three ice cubes sliding across the wood, macro lens, morning light.
Hands at work
Every model on this prompt →›What we asked for
Close-up of hands pouring hot water from a gooseneck kettle in slow circles into a pour-over coffee dripper, steam rising, shallow depth of field, warm cafe light.
Busy scene
Every model on this prompt →›What we asked for
A crowded Saturday farmers market seen from slightly above, shoppers with tote bags, a dog on a leash, a busker with a guitar, bunting swaying, late morning light, gentle camera drift.
Specific art style
Every model on this prompt →›What we asked for
A lighthouse on a rocky cliff in a storm, in the style of a Japanese ukiyo-e woodblock print, waves crashing in flat colour with visible wood grain, animated like moving paper cutouts.
Image to video
Every model on this prompt →›What we asked for
The coffee dripper slowly fills as coffee drips through it, steam rising, the camera drifting gently around it, the dripper itself unchanged.
Frequently asked questions
Is Sora 2 or Veo 3 better?
On our nine prompts Veo 3.1 wins on text, speech and physics, and Sora 2 is competitive on product and crowd shots. Sora 2's API is being removed on 24 September 2026, which settles it for anything new.
Is Veo 3.1 more expensive than Sora 2?
The full model is: $0.40 a second with audio against Sora 2's $0.10 at 720p. Veo 3.1 Fast at $0.15 and Lite at $0.05 bracket Sora's price on either side.
What replaces Sora 2?
Veo 3.1 Fast is the closest on price and audio. Kling 3.0 for longer clips, Seedance 2.0 for cinematic audio-first work. Each has a page here with the same nine clips.
Voyager
Run any of these inside Voyager
Voyager is an agent for creative work that runs every model on this page. Brief it, and it picks the model for the job, makes the thing and hands back something you can edit. Create an account to get started.