Video generation

Adobe Firefly Video Model

Name: Adobe Firefly Video Model
Author: Adobe

The Adobe Firefly Video Model generates video from a text prompt, optionally guided by an image used as the first or final frame.

A video generation model by Adobe · read 2026-08-14 · sourcing medium

Up to: 5 seconds
Resolution: 1920x1080 via the Firefly Services API (see disagreements: adobe.com markets 4K in the Firefly app)
Aspect ratios: 16:9, 9:16, 1:1
Weights: Not released
API: Yes
Commercial use: Vendor states commercial use is permitted

Adobe's Firefly Services API exposes it as model version video1_standard at POST /v3/videos/generate, described in Adobe's own reference as generating a five second video, with videoSettings fields for cameraMotion, shotAngle, shotSize and promptStyle. Its main selling point is provenance: Adobe says the model is trained on licensed content such as Adobe Stock plus public-domain material and not on Adobe user content, and that its output is designed to be safe for commercial use.

Commercial use — Vendor states commercial use is permitted

Adobe states the Firefly Video Model is trained on licensed content (e.g. Adobe Stock) and public-domain material, is not trained on Adobe user content, and that its output is designed to be safe for commercial use.

The licence notes recorded during research were read for a commercial-use position and did not state one. Read the licence before you rely on any of this. Nothing on this page is legal advice, and this site does not tell you a model is cleared for commercial use.

Specification

Vendor figures, read 2026-08-14Read from Adobe's own documentation on 2026-08-14. Vendors change these between model versions without announcing it, so check theirs before you commit a shot to a number. Sourcing for this entry is mixed. Some fields came from pages we retrieved, others could not be confirmed and are left blank. Check the vendor's current documentation

What it takes in: text, image
What it makes: video
Maximum duration: 5 seconds
Maximum resolution: 1920x1080 via the Firefly Services API (see disagreements: adobe.com markets 4K in the Firefly app)
Aspect ratios: 16:9, 9:16, 1:1
Controls: image-to-video, start-end-frame, camera-control
Weights: Not released
API: Yes
Pricing: Consumer access is via Adobe generative credits with a free daily allotment; Adobe does not publish a per-generation price for the Firefly Video Model on the pages reviewed, and Firefly Services API list pricing was not retrieved. Recorded as unverified.
Released or updated: model version video1_standard current in the Firefly Services API reference as of 2026-08-14; release date not stated on the pages reviewed

Strengths

Clear commercial-safety and training-data position, backed by Adobe indemnity positioning
Explicit cinematography parameters (camera motion, shot angle, shot size) in the API
First or final frame conditioning
Sits inside Premiere Pro, Express and Photoshop workflows
Same workspace lets you A/B the same prompt across Veo 3.1, Luma Ray3, Kling 3.0 and Runway Gen-4.5

Weaknesses

Five seconds per generation via the API, well short of Kling, Vidu and MiniMax
API resolution enum tops out at 1920x1080 despite 4K marketing claims
No native audio from the Firefly Video Model itself; Adobe points to partner models for that
Per-generation pricing is opaque (generative credits rather than a published rate)

Best for

Brand and marketing teams that need defensible provenance and Creative Cloud integration more than record clip length.

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.

{'claim': 'Maximum resolution and clip length', 'a': "Adobe's Firefly Services API reference describes 'Generate a five second video using a text prompt' and its usage notes list supported video dimensions only up to 1920x1080", 'aUrl': 'https://developer.adobe.com/firefly-services/docs/firefly-api/getting-started/usage-notes/', 'b': "adobe.com states 'Adobe Firefly delivers high-fidelity, 4K video generation' and that videos are downloadable in resolutions up to 4K, with clips up to 8 seconds on most models and 15 on some — but those statements cover the whole Firefly workspace including partner models, not the Firefly Video Model specifically", 'bUrl': 'https://www.adobe.com/products/firefly/features/ai-video-generator.html'}

What to understand before using it

Explainers naming the mechanisms this model's own documentation mentions.

Conditioning and control: everything you hand the model besides noise names “conditioning”
Text encoders and conditioning: how a prompt steers the picture names “conditioning”
Training data and copyright: what is not settled names “training”
Camera control: a move without a camera names “control”

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

Allegro

A small text-to-video model comprising a 2.8B VideoDiT and a 175M VideoVAE, released under Apache-2.0 in October 2024. It generates 88-frame clips at

Video generation

Amazon Nova Reel 1.1

Amazon Nova Reel generates video from text prompts or from text plus an image, and is available only through Amazon Bedrock. AWS documents output at 1

Video generation

ByteDance Seedance 2.0

Seedance 2.0 is ByteDance Seed's unified multimodal audio-video model: a single generation pass accepts text plus mixed references (up to 9 images, 3

Video generation

Sources

Read on 2026-08-14. A source that has changed since then will not be reflected here until the next pass.