Mirage / Captions
Mirage is a family of end-to-end audio and video foundation models that generate performances, voiceover, captions, graphics and music score, with the output kept as editable layers rather than a flat render.
- Weights
- Not released
- API
- Yes
- Commercial use
- Permitted subject to conditions
Recent work includes Avatar X for full-performance video generation and audio-driven avatar video. Captions is the consumer app (iOS, web and desktop) built on these models, offering AI Edit styles, digital twins and custom AI actors, video translation, AI voiceovers, B-roll generation and captions in 100+ languages; Mirage's APIs power bulk video generation and captioning.
Commercial usage terms not confirmed.
- Requires consent from the person depicted. Digital twins and custom AI actors are offered, but consent and likeness requirements were not stated on the pages fetched
Specification
- What it takes in
- video, image, audio, text
- What it makes
- video, audio, text
- Maximum duration
- not established
- Maximum resolution
- not established
- Aspect ratios
- not established
- Controls
- avatar-from-photo, avatar-from-video, lipsync, voice-clone, translation, auto-captions
- Weights
- Not released
- API
- Yes
- Pricing
- Credit subscription (iOS pricing shown): free tier with limited functionality, Max $24.99/mo for 500 credits, Scale 1x $69.99/mo for 1,400 credits, Scale 2x $139.99/mo for 2,800 credits, Scale 4x $279.99/mo for 5,600 credits, Enterprise custom with bulk credit discounts. Credits roll over up to 3x the monthly allowance; yearly billing is discounted.
- Released or updated
- not established
Maximum duration, maximum resolution and supported aspect ratios are not established here. That means no page we could retrieve stated them, not that the model lacks them.
Strengths
- Layered, still-editable output rather than a baked render
- Own foundation models across audio and video rather than reselling third-party APIs
- API for bulk generation and captioning
- Credit rollover up to 3x monthly allowance
Weaknesses
- No published resolution, duration or aspect ratio specs on any page fetched
- API documentation not publicly discoverable (docs.captions.ai does not resolve)
- Consent and commercial-use terms for digital twins undocumented on public pages
- Prices shown are iOS-only, so web/desktop pricing may differ
Best for
- High-volume short-form and UGC-style ad production
- Mobile-first creator workflows
- Bulk captioning and localisation
Where the sources disagree
Recorded rather than resolved. Picking the tidier number would hide the fact that the vendor's own pages do not agree.
- Brand structure is in flux: captions.ai/mirage 301-redirects to mirage.app/research, and the legal entity is NOCAP, Inc. It is unclear whether 'Mirage' is now the primary company brand with Captions as a product, or two brands in parallel. The research pages name Avatar X and reference 'Seeing Voices' and 'Shaping Voices' papers but publish no resolution, duration or release-date specs. docs.captions.ai does not resolve in DNS despite the site advertising APIs.
What to understand before using it
Explainers naming the mechanisms this model's own documentation mentions.
See it from the vendor
Shotcraft hosts no sample output. Judging a model means watching its own showreel with your own eyes, and a clip re-hosted here would be a clip we chose.
Nearby in the directory
Flawless DeepEditor and TrueSync
Neural performance editing tools for film and episodic post. DeepEditor applies new dialogue to existing shots and refines performances, covering in-v
Lip sync and avatars
Hedra character / avatar platform
Character and avatar video platform that has grown into a unified inference API putting multiple leading image, video and audio models behind one API
Lip sync and avatars
HeyGen Avatar IV (and Avatar V)
Avatar IV is HeyGen's current default avatar engine, driving a photo, consented digital twin or studio avatar from either a typed script (text to spee
Lip sync and avatars