Sora 2

Sora 2 is OpenAI's video model, launched on 30 September 2025 with synchronised audio and a social app, and it is on its way out. The consumer app closed on 26 April 2026, and on 24 March 2026 OpenAI told developers the Videos API and every Sora 2 model alias would be removed on 24 September 2026, with no replacement named. On the API it costs $0.10 a second at 720p; fal hosts it but marks the model unsupported upstream, which is why the samples below are what they are.

Anvisha Pai

Anvisha Pai, Co-founder & CEO, Voyager

Tested

OpenAI · released

Sora 2 filming itself as whimsical boardwalk scene, generated by us.
›What we asked for

A self-portrait of Sora 2, OpenAI's video model, as a whimsical scene: a golden retriever in sunglasses skateboarding down a Venice Beach boardwalk at sunset, onlookers cheering, synced skate wheel sounds and a distant guitar.

fal-ai/sora-2/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 4
returned 1280 by 720 pixels

At a glance

Retirement
App closed 26 April 2026. API, sora-2, sora-2-pro and every dated snapshot removed 24 September 2026.
Price
$0.10 a second at 720p for sora-2; sora-2-pro $0.30 to $0.70 by resolution.
Length
4, 8, 12, 16 or 20 seconds on fal's endpoint; our five-second fixture rounds down to four.
Audio
Native, synchronised.
Status on fal
Listed as no longer supported upstream, with elevated latency and timeouts.

Anvisha's take

Sora 2 worked on the day, which fal's own listing did not promise, and every clip came back in one to three minutes at $0.40 for four seconds. The output is fine and no more: the title card is set in a thin face that wobbles and truncates by the last frame, the speaking woman barely moves, and the spill is competent. Its self-portrait, a golden retriever skateboarding in sunglasses at sunset, is the most fun clip in the whole set. None of this matters for long: the API goes away on 24 September 2026.

Sora 2 real output: the same 9 clips every model gets

Every video model on this site gets the same 9 clips, so you can compare like with like. These are the clips Sora 2 returned, one try each, nothing re-rolled or retouched. Each tile shows one frame of the clip until you press play. Generated on . The 9 clips took 67 to 200 seconds each and cost $3.60 in total.

Product turntable4s clip · 77s wait · $0.40

A clean product shot that holds its shape through a full slow rotation.

›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.

fal-ai/sora-2/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 4
returned 1280 by 720 pixels

Person speaking4s clip · 87s wait · $0.40

A face, lip sync and a spoken line, where the model makes audio.

›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.

fal-ai/sora-2/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 4
returned 1280 by 720 pixels

Animated title card4s clip · 95s wait · $0.40

Set text that stays spelled correctly while it moves.

›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.

fal-ai/sora-2/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 4
returned 1280 by 720 pixels

Camera move4s clip · 67s wait · $0.40

A steady drone push-in with stable geometry.

›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.

fal-ai/sora-2/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 4
returned 1280 by 720 pixels

Physics4s clip · 87s wait · $0.40

A spill in slow motion: does water behave like water.

›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.

fal-ai/sora-2/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 4
returned 1280 by 720 pixels

Hands at work4s clip · 146s wait · $0.40

Hands, a kettle and a steady pour without extra fingers or teleporting objects.

›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.

fal-ai/sora-2/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 4
returned 1280 by 720 pixels

Busy scene4s clip · 146s wait · $0.40

Many people and objects moving at once without merging.

›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.

fal-ai/sora-2/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 4
returned 1280 by 720 pixels

Specific art style4s clip · 200s wait · $0.40

A named style held across every frame rather than drifting to photoreal.

›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.

fal-ai/sora-2/text-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 4
returned 1280 by 720 pixels

Image to video4s clip · 116s wait · $0.40

Bring a still product photo to life without changing the product.

›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. (starting from The Nano Banana Pro product shot from the image fixture set.)

fal-ai/sora-2/image-to-video
aspect_ratio: 16:9, resolution: 720p, duration: 4
returned 1280 by 720 pixels

How the cost under each picture was worked out: fal's Sora 2 rate of $0.10 per second at 720p (fal model page, read 2026-09-09) for the 4 seconds requested; fal does not return the charge.

Compare Sora 2

Sora 2 pricing

WhereUnitPriceSource
OpenAI APISecond at 720p, sora-2
sora-2-pro is $0.30 at 720p, $0.50 at 1024p, $0.70 at 1080p. Batch halves every rate. All of it ends 24 September 2026.
$0.10OpenAI API pricing, read
falSecond at 720p$0.10fal model page, read

Is Sora 2 free?

There is no free access to Sora 2 any more. Free ChatGPT users lost Sora generation on 10 January 2026, the Sora app closed on 26 April 2026, and the API has never had a free tier.

  • •Before January 2026 every tier had about 30 free generations a day; from 10 January it was Plus and Pro only.
  • •The Sora app shut on 26 April 2026.
  • •API: paid per second until 24 September 2026, then removed.

OpenAI API deprecations, read

Where to use Sora 2

  • OpenAI API · Model id sora-2 on the Videos API, until 24 September 2026.
  • fal · Hosted, marked unsupported upstream.
  • Azure AI Foundry · Microsoft's hosting of the same model, subject to the same retirement.

Voyager

Run Sora 2 inside Voyager

Voyager is an agent for creative work that runs every model in this reference, Sora 2 included. Brief it, and it picks the model, makes the thing and hands back something you can edit. Create an account to get started.

What we noticed running Sora 2

  • •The title renders in a thin face that wobbles, and by the last frame 'SESSIONS' has become 'SESS' and the date has lost letters. See the animated title card clip.
  • •The speaker is almost static: a small head turn and a smile, with the least motion of any clip on the roster for this prompt. See the person speaking clip.
  • •Self-portrait: The self-portrait came back as a skateboarding golden retriever in sunglasses on a boardwalk at sunset; the most cheerful clip in the set.
  • •Only 4, 8, 12, 16 or 20 second clips, so the five-second fixture is a four-second clip here. fal marks the model unsupported upstream; it ran anyway.

Sora 2 limits and API parameters

LimitValue
RetirementRemoved from the API on 24 September 2026, announced 24 March 2026, no replacement listed.
OpenAI API deprecations, read
Resolution720p only on sora-2: 1280 by 720 or 720 by 1280.
OpenAI model page, sora-2, read
Rate limits25 requests a minute at tier 1, rising to 375 at tier 5.
OpenAI model page, sora-2, read
›Every setting, for developers

fal (fal-ai/sora-2/text-to-video)

fal API reference, read

ParameterTypeValuesDefault
promptstringAny text
Required.
durationenum4, 8, 12, 16, 204
resolutionenum720p720p
aspect_ratioenum16:9, 9:1616:9
modelenumsora-2, sora-2-2025-12-08, sora-2-2025-10-06sora-2
character_idslistUp to two character references

What Sora 2 will not make

OpenAI's usage policies apply and every prompt and frame passes moderation; the Sora app additionally blocked real people without a consented cameo and applied a visible watermark that only Pro subscribers could remove. With the API's retirement on 24 September 2026 the practical policy question is moot: nothing new can be made with it after that date.

OpenAI API deprecations, read

Alternatives to Sora 2

Frequently asked questions

Is Sora 2 shutting down?

Yes. The Sora app closed on 26 April 2026, and OpenAI announced on 24 March 2026 that the Videos API and every Sora 2 model alias would be removed on 24 September 2026, with no replacement listed.

Is Sora 2 available through the API?

Until 24 September 2026, as sora-2 and sora-2-pro on the Videos API at $0.10 to $0.70 a second. After that, no.

How much does the Sora 2 API cost?

$0.10 a second at 720p for sora-2 and $0.30 to $0.70 a second for sora-2-pro, read on 9 September 2026, until the shutdown.

What should I use instead of Sora 2?

Veo 3.1 Fast is the closest on price and audio; Kling 3.0 Standard for longer clips; Seedance 2.0 for audio-first cinematic work. Each has a page here with the same nine prompts.

How this page is made

Prices, limits and settings are read from the linked vendor and host pages on the dates shown. Every picture was generated by us through fal with the request shown under it, and the original files are kept unedited. A generation that failed is shown as a failure. Model and vendor names are used to identify the products; no vendor imagery appears on this page.

Anvisha Pai

Anvisha Pai

Co-founder & CEO, Voyager

Anvisha is the CEO of Voyager and a repeat, Y Combinator-backed startup founder. She was previously a PM at Dropbox. She believes nobody should need a design degree to make something that looks great.