Structured model database
AI video models
Find newly released and recently updated models, then filter public records by capability, company, country, access, and evidence.
13public models6populated capability views
Latest models
Newest verified releases
Ordered only by source-verified release date. Unknown dates appear under Recently updated.
P-Video-2-Pro
Pruna AI
- Released
- Sep 17, 2026
- Max duration
- 5–15 seconds
- Max resolution
- 768P
- Native audio
- Yes
View model
Vidu S2
ShengShu Technology / Vidu
- Released
- Sep 15, 2026
- Max resolution
- 720p
- API
- Available
View model
SwitchX 2.0
Beeble AI Inc.
- Released
- Sep 9, 2026
- Max duration
- Up to 20 seconds / 480 frames at 24 fps
- Max resolution
- Up to 2160p on Beeble Cloud
- API
- Available
View model
Gemini Omni Flash 1.1
- Released
- Aug 27, 2026
- Max duration
- 3–10 seconds per generated clip; extension can continue in additional steps
- Max resolution
- 4K delivered; 720p native/default; 1080p and 4K upscaled
- Native audio
- Yes
View model
H3 Max
fal Research / MiniMax
- Released
- Aug 26, 2026
- Max duration
- 5–15 seconds
- Max resolution
- 768P native; 1080P latent refinement on base endpoints
- Native audio
- Yes
View model
Wan 3.0
Alibaba Cloud / Alibaba
- Released
- Aug 6, 2026
- Max duration
- 2–30 seconds
- Max resolution
- 1080P (official Alibaba ceiling)
- Native audio
- Yes
View model
Seedance 2.5
ByteDance Seed
- Released
- Jul 31, 2026
- Max duration
- 30 seconds per generation
- Native audio
- Yes
- Image to video
- Yes
View model
Grok Imagine Video 1.5
xAI / SpaceXAI
- Released
- Jun 16, 2026
- Max duration
- 1–15 seconds canonical xAI generation; longer Roko buckets are provider-specific and non-official
- Max resolution
- 1080p native for T2V/I2V; R2V max 720p; edit inherits input and is capped at 720p
- Native audio
- Yes
View model
Freshness
Recently updated
Records with the newest content or verification activity.
Browse by capability
Start with a verified feature
These controls filter the explorer; they do not create standalone SEO URLs.
Full explorer
Filter every public model
Unknown fields remain visible instead of silently becoming a “no.”
13 results
| Model | Best for | Starting price | Native audio | Max resolution | API | Commercial use | Our score |
|---|---|---|---|---|---|---|---|
| Gemini Omni Flash 1.1Google | Native-audio text, image and reference-to-video workflows | Yes | 4K delivered; 720p native/default; 1080p and 4K upscaled | available | Pending | ||
| Grok Imagine Video 1.5xAI / SpaceXAI | Native-audio text, image and reference-to-video workflows | Yes | 1080p native for T2V/I2V; R2V max 720p; edit inherits input and is capped at 720p | available | Pending | ||
| H3 Max Turbofal Research / MiniMax | Fast hosted text-to-video and image-to-video iteration | From $0.0125 | Yes | Up to 1080P; endpoint tiers are 480P, 768P, and 1080P | available | fal marks the hosted endpoint for commercial use; review fal’s terms for the applicable license conditions. | Pending |
| HappyHorse 1.1Alibaba / Alibaba-ATH | Native-audio text and image to video | Yes | 1080P | available | Pending | ||
| InVideoInVideo | End-to-end production | Free / open access | unknown | Pending | |||
| H3 Maxfal Research / MiniMax | Fast hosted text-to-video iteration | Yes | 768P native; 1080P latent refinement on base endpoints | available | true | Pending | |
| P-Video-2-ProPruna AI | Fast hosted T2V/I2V with native generated audio | Yes | 768P | available | Pending | ||
| Seedance 2.5ByteDance Seed | Long-form narrative clips | Yes | announced | Pending | |||
| SwitchX 2.0Beeble AI Inc. | Video-to-video VFX on existing source footage | Up to 2160p on Beeble Cloud | available | Beeble Cloud subscription plans include a full commercial license; API terms are separate and were not verified in this intake. | Pending | ||
| SynthesiaSynthesia | Training videos | Free / open access | available | Pending | |||
| Q3 TurboShengShu Technology / Vidu | Native audio text-to-video and image-to-video workflows | Yes | 1080p canonical; provider-only 360p is not a Vidu model tier | available | Pending | ||
| Vidu S2ShengShu Technology / Vidu | Real-time interactive avatars | 720p | available | Pending | |||
| Wan 3.0Alibaba Cloud / Alibaba | Mode-aware text and image to video evaluation | Yes | 1080P (official Alibaba ceiling) | available | Pending |