Runway vs Veo for Video Ads: The Same $0.40 a Second, Bought Two Ways

Eight seconds with sound: $3.20 from Google or Runway's API

A growth lead at a coffee brand has two browser tabs open on a Monday. One is Google's Gemini API pricing page, where her developer already has a key. The other is Runway, where the design team already pays for seats. Both sell the same model. She needs twelve eight-second vertical hooks with a spoken line, and she wants to know which tab should get the order.

Price the obvious cell and the tabs agree. As of September 2026, Google's Gemini API lists Veo 3.1 at $0.40 per second of video with audio, at 720p and at 1080p. Runway's developer API bills Veo 3.1 with audio at 40 credits a second and sells credits at $0.01 each, which is also $0.40. An eight-second clip comes to $3.20 at either counter.

So "Runway vs Veo" is not really a fight between two models. Veo 3.1 is Google's model, and Runway is one of the places that sells it, next to its own Gen-4.5 and Aleph 2.0. The useful questions are narrower: where to buy the Veo seconds, and whether you want Runway's suite wrapped around them. One disclosure before the numbers: Novoads competes in AI video ads and runs Google Veo 3.1 as well, so we have a stake here. Every rival price below comes from the vendor's own page, read on September 17, 2026.

The short answer: which counter fits which job

The two counters split on four things: the cheaper tiers, silent clips, who writes the code, and what happens to the clip after it renders.

Buy Veo 3.1 from Google when

• Someone on the team can call an API. Google sells Veo through the Gemini API, and its Veo page also points to the Gemini app and Google Flow.
• You want the cheapest clips with sound. Veo 3.1 Fast lists at $0.10 per second at 720p, and Veo 3.1 Lite at $0.05.
• You need 4K. Google lists Standard at $0.60 per second in 4K and Fast at $0.30.
• You want failed renders off the bill. Google's pricing page says, "You will only be charged if your video is successfully generated."

Buy Veo 3.1 through Runway when

• You would rather not manage a second vendor. Runway says you use Veo 3.1 with your Runway plan, "with no separate Google account or subscription required."
• The clip is silent. Runway bills Veo 3.1 without audio at 20 credits a second, half its rate with sound.
• The footage gets edited in the same place. Aleph 2.0 edits, Studio assembly and an Adobe panel sit on the same plan.
• Some shots don't need Veo at all. Runway's own Gen-4.5 runs on the same credits for visual shots with no voice.

Edge on the cheapest sound-on tiers: Google. Edge on silent Veo and on finishing the ad: Runway.

What a "Runway vs Veo" search really compares

Most searches for this matchup assume two rival generators. In practice, one is a store and the other is a product on its shelf.

Veo 3.1 is Google's model, sold at more than one counter

Google DeepMind calls Veo 3.1 "our leading video generation model, designed to empower filmmakers and storytellers." On the Gemini API it runs under three preview model IDs, one per tier:

• veo-3.1-generate-preview for Standard.
• veo-3.1-fast-generate-preview for Fast.
• veo-3.1-lite-generate-preview for Lite, which Google's changelog says it launched on March 31, 2026.

The preview label matters for planning. Google's own changelog points teams that need the generally available versions elsewhere, to "the 3.1 GA models available through the Gemini Enterprise Agent Platform." The per-second prices in this post are the Gemini API's.

Runway is a creative suite that stocks Veo

Runway describes its lineup as "Runway models like Gen-4.5 and Aleph 2.0, plus third-party models like Nano Banana Pro 2, GPT Image 2, Kling 3.0, Kling O3, Veo 3.1, Seedance 2.5 and more." Its Veo page is plain about access: "Veo 3.1 is available on all paid and Enterprise plans. Free plan users can upgrade to get access."

Developers get Veo from Runway too. Runway's API price list carries two Veo 3.1 models, veo3.1 and veo3.1_fast, each priced with audio and without. Runway also lists Veo 3.1 among the models its MCP connector can reach from an AI agent.

Two matchups hide inside one search

1. Model against model. Veo 3.1 against Runway's own Gen-4.5, for the shot itself.
2. Counter against counter. Google's API against Runway's plans and API, for the very same Veo 3.1 clip.

The first is mostly a question about sound, which the next section takes on. The second is a pricing question, and it gets two sections of its own. For the wider field around Runway, our guide to Runway alternatives for ad creators compares six other tools. If Veo's price is the problem rather than the counter, our Veo 3 alternatives guide maps the cheaper models to the ads each one makes.

Sound is what separates Veo 3.1 from Gen-4.5

A UGC-style ad is usually a person talking to camera. That makes audio the first spec to check, and it is where Veo 3.1 and Runway's own model part ways.

Veo 3.1 renders the voice in the same pass

Google says Veo lets you "add sound effects, ambient noise, and even dialogue to your creations." On the Gemini API, the Veo feature table marks audio as "always on" for Veo 3.1, Veo 3.1 Fast and Veo 3.1 Lite. The spoken line, the room tone and the effects come back with the picture, so there is no dubbing step to drift out of sync.

Language is the caveat for global ad teams. Google's Veo guide says "English (en) is fully supported, but other languages have not been evaluated, so they may work but results can vary." A Spanish or German hook deserves a test render before you queue fifty of them.

Runway sells a silent Veo at half the credits

Runway treats Veo's audio as a switch. Its Veo page tells you to generate with native audio "or without it when you're scoring the edit yourself," and its credits page prices the two cases apart:

• Veo 3.1 with audio: 40 credits per second.
• Veo 3.1 without audio: 20 credits per second.

Runway's API reference says the audio setting defaults to on and adds, "Audio inclusion affects pricing." For a pour, a product spin, or B-roll that will sit under a voiceover recorded elsewhere, the silent setting halves the bill.

Gen-4.5 is Runway's own visual lane

Runway pitches Gen-4.5 on motion, saying it "achieves unprecedented physical accuracy and visual precision." At 12 credits a second it costs less than even silent Veo 3.1, and on Runway's API a Gen-4.5 clip runs 2 to 10 seconds. Runway's plans list music, voice and sound effects generation as a separate feature, so a spoken ad built on Gen-4.5 takes a second step for the voice. The practical split for an ad team:

• Spoken hooks and testimonials: Veo 3.1, or another audio-native model on the same shelf.
• Silent product shots and B-roll: Gen-4.5, or silent Veo 3.1 when you want Veo's look.

What blind votes settle, and what they leave open

On Artificial Analysis' text-to-video leaderboard with audio, fetched September 17, 2026, Veo 3.1 scores 1088 Elo, Veo 3.1 Fast 1085 and Veo 3.1 Lite 1083, each within a margin of about seven points. Two things stand out for a buyer. The three Veo tiers sit within a few points of each other on that board, while Lite costs an eighth of Standard per second at 720p. And Google's other video model, Gemini Omni Flash, leads the same board at 1233. Runway sells it on its API too, at 10 credits a second for text to video.

Runway's own chart for Gen-4.5, with 1,247 Elo points, is a snapshot dated November 30, 2025, so it cannot be set against those September numbers, the same caveat our Runway vs Seedance comparison makes. Neither ranking tells you whether your product label survives the render. That is a job for your own test, and part of why realism is a tie between tools renting the same engines, as we explain in where AI UGC realism actually comes from.

Price per second at each counter

Every surface in this comparison meters Veo differently: Google in dollars per second, Runway's API in credits per second, Runway's plans in monthly credit pools. Converting all of it to dollars for an eight-second clip is the only comparison that holds, which is the same logic our explainer on AI video credits walks through.

Google's Gemini API list prices

Google's pricing page, last updated September 16, 2026, lists these paid-tier prices per second for video with audio:

• Veo 3.1 Standard: $0.40 at 720p and 1080p, $0.60 at 4K.
• Veo 3.1 Fast: $0.10 at 720p, $0.12 at 1080p, $0.30 at 4K.
• Veo 3.1 Lite: $0.05 at 720p, $0.08 at 1080p, with no 4K output.

The free tier column reads "Not available" for all three. These are not new prices. Google launched Lite on March 31, 2026, and said in the same announcement, "On April 7, we'll also be reducing the pricing for Veo 3.1 Fast." The same page adds a warning worth reading before a big batch: "an audio processing issue may prevent a video from being generated," and only successful videos are charged.

Runway's API: the same Standard price, plus a silent row

Runway Dev sells credits at $0.01 each, and prices its two Veo models like this:

• veo3.1 with audio: 40 credits per second, or $0.40.
• veo3.1 without audio: 20 credits per second, or $0.20.
• veo3.1_fast with audio: 15 credits per second, or $0.15.
• veo3.1_fast without audio: 10 credits per second, or $0.10.

Runway's price rows do not name a resolution. Its API reference accepts four output ratios for veo3.1: 1280:720, 720:1280, 1080:1920 and 1920:1080. Our reading is that one per-second price covers 720p and 1080p in both orientations, which makes the Standard row a like-for-like match with Google's $0.40. Clips run 4, 6 or 8 seconds on both counters.

Eight-second clip	Google Gemini API	Runway API
Standard, sound on	$3.20	$3.20
Standard, sound off	audio always on	$1.60
Fast, sound on, 720p	$0.80	$1.20
Fast, sound on, 1080p	$0.96	$1.20
Fast, sound off	audio always on	$0.80
Lite, sound on, 720p	$0.40	no Lite row

As of September 17, 2026, Runway's API price list has no Lite row, so the cheapest Veo clip with sound is Google's.

Runway's web plans: the same clip in credits

Runway's credits page lists Veo 3.1 at 40 credits a second with audio, so an eight-second clip with sound uses 320 credits, and a silent one 160. Here is what each monthly allowance holds if every credit goes to Veo:

• Standard: $15 a month, or $12 billed annually, for 625 credits. One clip with sound, or three silent.
• Pro: $35 a month, or $28 annually, for 2,250 credits. Seven with sound, or fourteen silent.
• Max: $95 a month, or $76 annually, for 9,500 credits. Twenty-nine with sound, or fifty-nine silent.

The plan-rate formula: plan price × 320 ÷ plan credits = dollars per eight-second Veo 3.1 clip with sound. Billed monthly, that is $7.68 on Standard, $4.98 on Pro and $3.20 on Max. Billed annually, Max falls to $2.56, below Google's Standard list price. The catch is use-it-or-lose-it: on Standard and Pro, monthly credits don't roll over, while Max carries up to one month forward.

A worked example: twelve hooks and one hero cut

Back to the coffee brand. The brief is twelve eight-second vertical hooks, each with one spoken line, for a Reels test, plus a 1080p hero version of the winner. That is thirteen clips and 104 seconds of video with sound, before a single retake.

The brief, priced at each counter

Route	12 hooks	Hero clip	Total
Google, Standard hooks	$38.40	$3.20	$41.60
Google, Fast 720p hooks	$9.60	$3.20	$12.80
Google, Lite 720p hooks	$4.80	$3.20	$8.00
Runway API, veo3.1 hooks	$38.40	$3.20	$41.60
Runway API, Fast hooks	$14.40	$3.20	$17.60
Runway plan, Veo 3.1	3,840 credits	320 credits	4,160 credits

On Runway's web plans, 4,160 credits is more than Pro's 2,250 for a month, so the batch needs Max, or Pro plus extra credits. Runway sells those in a minimum purchase of 1,000, and credits you buy on top of a plan never expire.

The arithmetic, in plain terms

At full Standard quality the two counters cost the same, $41.60. Drop the hooks to Google's Fast tier and the brief costs less than a third of that. On Runway's API, the same move saves less, because Fast with audio is 15 credits there against Google's $0.10. The hero clip is $3.20 on both APIs, and it has to be eight seconds long, because Google requires eight-second clips for 1080p and 4K output.

What the arithmetic leaves out

• Takes you throw away. Google charges only for successful generations, but a successful take with a garbled product label still bills at every counter. Budget retakes into the hook tier, not the hero.
• Waiting. Google's Veo guide lists request latency from 11 seconds to 6 minutes during peak hours. Our own measured render times show why a batch plan needs slack.
• Storage. On the Gemini API, "generated videos are stored on the server for 2 days, after which they are removed." Download the winners the day they render.
• Reading the test. Twelve hooks only pay off if you read the result honestly, which our guide to ad creative testing covers.

Control and limits at each counter

Price is only part of the purchase. Each counter exposes a different set of controls, and a few of them change what an ad team can make.

What Google's API exposes

Google's Veo guide documents more control than a typical app surface shows:

• Reference images: "Veo 3.1 now accepts up to 3 reference images to guide your generated video's content," with images of "a person, character, or product" named as the use.
• First and last frames: generate the motion between two pinned images.
• Extension: extend a Veo video "by 7 seconds and up to 20 times," to a combined output of up to 148 seconds, at 720p and not on Lite.
• Format: 16:9 by default or 9:16, with every clip watermarked by Google's SynthID.

The trade is that eight seconds is the working unit. The API requires an eight-second clip when you use extension, reference images, 1080p or 4K.

What Runway exposes around the same model

Runway's surfaces for Veo 3.1 are simpler and sit inside a larger tool:

• In the app: text prompts, images for image-to-video, and start and end frames, with the audio switch described above.
• On the API: a prompt of up to 1,000 characters, a first frame with an optional last frame, and clips of 4, 6 or 8 seconds.
• Through agents: Veo 3.1 appears among the models Runway's MCP connector can call.

Runway's API pricing does not show a 4K Veo rate, so for 4K Veo the Gemini API is the counter with a published price.

Where Runway finishes the ad

The case for buying Veo through Runway is what happens after the clip exists, the same editing edge our Luma AI vs Runway comparison gives Runway. Runway's Aleph 2.0 works on footage you already have: "Edit one frame and Aleph 2.0 modifies the rest of your video to match," on clips up to 30 seconds at 1080p, at 28 credits a second on the API. The rest of the finishing stack is dated in Runway's changelog:

• Studio, added June 18, 2026, to "Trim, stitch, reorder and export a final video."
• A panel inside Adobe Premiere Pro and After Effects, on paid plans since September 8, 2026.

Our Runway vs Kling comparison goes deeper on that editing workflow, and our Kling vs Veo breakdown sets Veo 3.1 against Kling, another model Runway sells with native audio and speech.

How Novoads solves the ad around a Veo clip

Novoads is a global AI UGC video-ad generator: you upload a product photo, write or auto-generate a script, pick an AI actor and export a vertical ad. Google Veo 3.1 runs inside Novoads next to Seedance 2.0, Seedance 2.0 Mini, Seedance 2.5, Kling v3 Pro and Omni Flash, which is Google's Gemini Omni Flash, all on one credit balance.

Veo 3.1 at a flat 10 credits per clip

In Novoads, a Google Veo 3.1 clip costs 10 credits whatever its settings:

• Length: 4, 6 or 8 seconds.
• Resolution: 720p or 1080p.
• Shape: 9:16 or 16:9.
• Sound: generated on every clip.

Novoads starts at $49/month (Starter, 50 credits per month), and all plans are published on /pricing. That makes a Veo 3.1 clip $9.80 of a Starter plan, or $7.90 on the $79 Pro plan with 100 credits. Here is the concession, stated plainly: that is about three times Google's $3.20 list price for the same eight seconds. If raw Veo seconds are all you need, buy them from Google.

What the credits buy instead

The case for Novoads is the ad around the clip, not a cheaper Veo second:

• Cheaper engines for testing on the same plan. Seedance 2.0 Mini starts at 1.5 credits for five seconds, so hooks can run there while Veo is saved for the finish. The thirteen-clip brief above would be 130 credits on Veo alone.
• The rest of the ad in one project. Scripts, AI actors, voices in 31 languages and captions, with no code to write.
• What Novoads does not do. It does not offer Runway's Gen-4.5 or Aleph 2.0, a silent Veo discount, or 4K Veo. If the job is editing footage you already shot, Runway is the better tool.

If the job is a steady run of spoken, product-first vertical ads, you can start on Novoads and pick Veo 3.1 for the cuts that earn it.

Veo is the same everywhere. The bill is not.

Runway already sells Veo 3.1, so this comparison was never really about which model makes the better clip. It is about the counter. Google sells the cheapest seconds with sound and publishes a 4K price. Runway matches Google's Standard rate, sells silent Veo at half price, and wraps the clip in an editing suite. Pick the counter by what happens after the render, because the model you buy is the same at both.