What actually makes up AI video ad cost
Short answer: on RelayMe a 4-second 720p ad clip with audio costs 480 to 600 credits (about US$0.48 to US$0.60) on the fast tiers — MiniMax-H3 at 120 credits per second, Veo 3.1 Fast or Grok Imagine 1.5 at 150 — and 1,600 credits (US$1.60) on the top-quality Veo 3.1 tier; 1,000 credits is roughly US$1. Add the AI product image that seeds the clip (20 credits) and a full image-to-video ad chain lands at about 620 credits, which is what our own August 2026 production test measured. The most cost-effective platform is the one with the lowest price per usable clip, not the lowest list price, which is why drafting on a cheap tier and finalizing on a quality tier inside one workflow matters more than any single model's rate.
The real cost of an AI video ad has four parts: model fees (usually tied to duration and quality tier), re-run waste (regenerating unsatisfying results), human time (prompting, screening, editing), and exploration cost (the batches it takes to find a usable style). Most teams underestimate the last three — the listed model price is only the starting point, and the effective cost per usable clip is often a multiple of it.
The first step to controlling spend is tracking these four buckets separately so you can see where the optimization headroom is. Model prices change frequently; always confirm current rates on each provider’s official page.
Cost per second by model (RelayMe pricing, September 2026)
Every video model on RelayMe bills per second of output. Veo 3.1 Fast: 150 credits per second at 720p or 1080p with audio, 100 without, 350 or 300 at 4K. Veo 3.1: 400 credits per second with audio, 200 without, 600 or 400 at 4K. Grok Imagine 1.5: 150 credits per second at 720p or 1080p with or without audio, no 4K tier. MiniMax-H3: 120 credits per second with audio at 720p or 1080p, 200 at 4K. Because 1,000 credits is roughly US$1, a with-audio second costs US$0.12 to US$0.40 depending on the model, and a silent draft second US$0.10 on Veo 3.1 Fast.
Translated into ad units: a 4-second 720p clip with audio is 480 credits on MiniMax-H3, 600 on Veo 3.1 Fast or Grok Imagine, 1,600 on Veo 3.1; an 8-second clip doubles those; at 4K with audio the same 4 seconds is 800 credits on MiniMax-H3, 1,400 on Veo 3.1 Fast and 2,400 on Veo 3.1. The US$19 Starter plan's 20,000 monthly credits therefore cover roughly 33 four-second Veo 3.1 Fast clips with audio, or 50 silent drafts, before re-runs.
Which is most cost-effective depends on what you are buying: MiniMax-H3 is cheapest per second and cheapest at 4K; Grok Imagine 1.5 charges the same with or without audio, so it wins for sound-led ads; Veo 3.1 Fast is the balanced default for drafts and volume; Veo 3.1 is worth its 2.7x premium only on hero shots where the quality gap is visible in the final ad. Comparing platforms, put the per-second with-audio price next to your observed re-run rate — a cheaper model that needs three attempts costs more than a pricier one that lands in one. Prices change frequently; the live table is on the RelayMe pricing page.
Check the current pricing and credit rules before selecting a model; a dated example is not a current quote.
How mainstream models price generation
Billing for mainstream video models such as Veo and Kling typically depends on three factors: clip duration, resolution tier, and generation mode (fast draft versus high quality). The same prompt can cost several times more after switching tiers.
Do not compare list prices alone — estimate by your real usage: iterate on low tiers during A/B testing, then render final cuts on high tiers. Since prices and tiers shift with releases, note the quote date for each model in your budget sheet.
For a concrete anchor: in our own August 2026 test, a 4-second 720p clip with audio on Veo 3.1 Fast consumed 600 credits (150 credits per second, about $0.60), and the product image driving it cost 20 credits on an image model — the whole image-to-video chain came to 620 credits, roughly $0.62. Newer models keep the same billing shape: Grok Imagine, added to RelayMe in late August 2026, bills a flat 150 credits per second, with or without audio.
Four cost-saving tactics that work
First, image before video: lock composition and style with low-cost image generation, then render video only after approval, cutting expensive re-runs. Second, template your prompts: save proven prompts as reusable templates and swap only the product variables for new items.
Third, split tiers: drafts on the fast cheap tier, finals on the high-quality tier. Fourth, track re-run rate: measure re-runs per asset type and fix high re-run prompts before paying for a pricier model. Teams that apply all four usually see the effective cost per usable clip drop substantially.
A fifth lever worth naming: model failover. In our August 2026 test the first video provider failed twice within seconds; switching the same node to Veo 3.1 Fast reused the prompt and start frame unchanged and succeeded on the first try. The failed attempts cost nothing, and the ability to change providers without rebuilding the flow is what keeps one bad provider day from turning into a budget problem.
Once a still image is approved, the image-to-video guide explains the next stage and its dated worked example.
Audit spend with task history
RelayMe keeps a task history of every generation, which lets you audit spend against output: which workflow has the highest usable rate, and which prompt patterns waste the most.
Review it weekly and build a simple report of clips generated, clips used, and re-runs. When an asset type stays low on usable rate, revisit the prompt and reference material before changing models or tiers.
Our own August 2026 run shows what that audit looks like line by line: 20 credits for the seed product image, 600 credits for the 4-second clip, and two failed attempts on another provider that cost nothing because the node was simply switched to a different model and rerun. That level of per-line visibility is what makes a weekly review take minutes instead of an afternoon.
Scale your video production workflow
Once a single flow is stable, freeze it as a RelayMe workflow template: a language model step expands product details into video prompts automatically, and the video step renders the result, so operators only maintain the product sheet.
At scale, the lever is no longer unit price but usable rate and screening time — both usually outweigh the price gap between models. Validate template stability on small batches before ramping volume.
Scale is also where per-second pricing becomes a planning number rather than trivia: at 150 credits per second on fast tiers, a monthly batch of 100 four-second clips budgets at roughly 60,000 credits — above the Basic plan's 35,000 monthly credits but comfortably inside Pro's 110,000, or coverable with never-expiring top-ups. Once you know your per-clip number, matching plan size to volume is five minutes of arithmetic.