Which should you pick?
Cosmos 3 Super Image to Video is the cheaper job: 8-28 credits against 32-128 credits for VEO First Last Frame. Those are the credits one complete generation costs, not a headline rate, so they are directly comparable even though the two models bill differently.
At their most expensive settings the gap widens: VEO First Last Frame tops out at 128 credits a job versus 28, driven by audio options.
VEO First Last Frame supports up to 4K, the higher of the two.
Cosmos 3 Super Image to Video supports more aspect ratios (5: 16:9, 9:16, 1:1, 4:3, 3:4), giving it more format flexibility.
What a job costs
VEO First Last Frame
32-128 credits
It bills by whether audio is generated, and the price moves with audio options.
- Audio off8 credits
- Audio on16 credits
Cosmos 3 Super Image to Video
8-28 credits
It bills per second of output, and the price moves with clip length.
- Price4 credits
All figures are Versely credits. Your plan's credit allowance is on the pricing page.
Spec comparison
| Spec | VEO First Last Frame | Cosmos 3 Super Image to Video |
|---|---|---|
| Provider | NVIDIA | |
| Cost per job | 32-128 credits | 8-28 credits |
| Billing | by whether audio is generated | per second of output |
| Max resolution | 4K | 720p |
| Aspect ratios | 16:9, 9:16, auto | 16:9, 9:16, 1:1, 4:3, 3:4 |
| Durations | 4s, 6s, 8s | 2s, 3s, 4s, 5s, 6s, 7s |
| Frame rates | 24fps, 30fps | — |
| Qualities | 720p, 1080p, 4K | 480p, 720p |
| Requires starting image | Yes | Yes |
| Accepts video input | No | No |
| Reference images | — | — |
| Leaderboard | #22 in image to video | Unranked |
| Released | Oct 15, 2025 | Aug 1, 2026 |
What each one does that the other doesn't
Only VEO First Last Frame
frame control, precise, premium
Only Cosmos 3 Super Image to Video
image driven, physics, prompt expansion, value, top ranked
Both list: image to video.
Leaderboard positions
Versely ranks each capability on its own leaderboard, so a position is only comparable between two models within the same row.
VEO First Last Frame
- Image to video#22
Cosmos 3 Super Image to Video
Not on a Versely leaderboard yet.
Full VEO First Last Frame spec sheet →
Generate videos between first and last frames
Full Cosmos 3 Super Image to Video spec sheet →
NVIDIA's Cosmos 3 Super world model animates a still image with physically grounded motion, guided by a text prompt that a built-in reasoner expands first. 480p or 720p, 2-7 seconds at 24 fps.
More comparisons
Frequently asked questions
Which is cheaper, VEO First Last Frame or Cosmos 3 Super Image to Video?+
Cosmos 3 Super Image to Video is cheaper per job: 8-28 credits against 32-128 credits for VEO First Last Frame.
What can VEO First Last Frame do that Cosmos 3 Super Image to Video can't, or vice versa?+
Only VEO First Last Frame lists frame control, precise, premium. Only Cosmos 3 Super Image to Video lists image driven, physics, prompt expansion, value, top ranked. Both list image to video.
How do VEO First Last Frame and Cosmos 3 Super Image to Video bill?+
VEO First Last Frame bills by whether audio is generated, and its price moves with audio options. Cosmos 3 Super Image to Video bills per second of output, and its price moves with clip length. Both are charged in Versely credits.
Which has higher output resolution, VEO First Last Frame or Cosmos 3 Super Image to Video?+
VEO First Last Frame supports higher output (4K against 720p).
How long a clip can VEO First Last Frame and Cosmos 3 Super Image to Video generate?+
VEO First Last Frame supports 4s, 6s, 8s; Cosmos 3 Super Image to Video supports 2s, 3s, 4s, 5s, 6s, 7s.
Try both inside Versely
The all-in-one AI studio for creators. 60+ models for video, image, voice, music and lipsync, in your browser or on your phone.
Free on the web and on iPhone, one account.