Audit a rival price-per-second table against the vendor pages
A Higgsfield post lists Veo 3.1 at $0.40/s and Runway Gen-4.5 at $0.12/s. Google matches the first; Runway plan math gives $0.096 to $0.23. Check yours.

Treat any vendor-authored price table as a lead, not a fact, and check each row on the seller's own page. A Higgsfield comparison dated September 2026 lists Veo 3.1 Standard at $0.40 per second, Runway Gen-4.5 at $0.12, Luma Ray 3.2 at $0.09, MiniMax H3 at $0.08 and Seedance 2.5 at $0.231 at 720p. Only some of those rows can be matched against first-party pages we read on 2026-10-03.
| Model | Higgsfield post | First-party page | Result |
|---|---|---|---|
| Veo 3.1 Standard | $0.40/s | $0.40/s at 720p and 1080p (Google) | Matches |
| Runway Gen-4.5 | $0.12/s | 12 credits/s; $0.096 to $0.23/s by plan (Runway pricing math) | Within range, plan dependent |
| Luma Ray 3.2 | $0.09/s | Credit range 4 to 314+ per second (Luma) | Cannot confirm without the model's rate |
| MiniMax H3 | $0.08/s | No first-party page read | Unverified |
| Seedance 2.5 at 720p | $0.231/s | No first-party page read | Unverified |
How the checks were done
Google's Gemini pricing page lists Veo 3.1 Standard at $0.40 per second for 720p and 1080p. For Runway, the pricing page gives Gen-4.5 at 60 credits per 5 seconds, which is 12 credits per second; plan credit prices of $0.0192, $0.0124 and $0.0080 give $0.230, $0.149 and $0.096 per second. The $0.12 row lands inside that range.
Rules for using rival tables
- Check the unit: per second, per clip or per credit.
- Check the plan or tier behind a credit price.
- Note who wrote the table; the Higgsfield post is vendor-authored and is itself a seller.
- Date every number and re-read before quoting it.
On Sume
Sume prices are provider list times 1.25, shown per model in pricing_skus from GET /v1/videos/models, so you can check them with one API call instead of trusting a table. See the video docs.
Sources
Related posts
More in Comparisons
- fal image model list vs the Sume catalog: how to check by query
fal.ai lists Seedream 5.0, GPT Image 2.5, Flux 2, Nano Banana 2, Ideogram 4 and Krea 2. How to see which ones a Sume key can call, with a short Python script.
- fal retry budgets and 1-hour grace vs Sume job error categories
fal's changelog lists per-condition retry budgets and termination grace of up to one hour. Sume job errors carry a category, a next action and retry-after.
- fal's Usage API vs Sume's per-job, per-run cost reads
fal's changelog lists a Usage API for cost attribution. Sume's GET /v1/usage filters by job_id, run_id or thread_id and returns a summary. How to attribute.
- FLUX.2 [klein] needs about 13 GB VRAM: local run or API?
Recraft cites about 13 GB of VRAM for FLUX.2 [klein] under Apache 2.0. A checklist for deciding between a local GPU and a hosted image API.
Written by Sume