AI Model Disclosure
FacelessPilot identifies the actual generation models used by its production workflow and standalone media tools. Model names identify third-party integrations only and do not imply sponsorship, endorsement, or ownership by the model developer.
Core faceless-video workflow
The standard Series and Episode production workflow currently uses these models:
| Feature | Model | Model identifier | API provider |
|---|---|---|---|
| Scene and cover image generation | GPT Image 2 | gpt-image-2-text-to-image | Kie API |
| Image editing and visual continuity | GPT Image 2 | gpt-image-2-image-to-image | Kie API |
| Optional 5-second motion intro | Grok Imagine Video 1.5 Preview | grok-imagine-video-1-5-preview | Kie API |
The standard finished video is assembled from approved images, narration, captions, and music using our server-side rendering pipeline. It is not represented as output from a separate text-to-video model. When the optional AI motion intro is enabled, only that opening segment is generated with Grok Imagine Video 1.5 Preview.
Standalone image generator
Where the corresponding provider is configured and available, the image model switcher may offer:
- GPT Image 2 through Kie API;
- Seedream 4.5 through fal;
- Nano Banana Pro (Gemini) through Replicate, fal, or Kie API;
- Seedream 4 through Replicate or fal;
- Z-Image Turbo through fal;
- Flux 2 Flex through fal; and
- Gemini 3 Pro Image Preview through Google Gemini.
The model selected in the interface is the model submitted for that generation request.
Standalone video generator
Where the corresponding provider is configured and available, the video model switcher may offer:
- Veo 3.1 through Replicate;
- Sora 2 through Replicate;
- Veo 3 through fal;
- Wan Pro through fal;
- Kling Video O1 through fal; and
- Grok Imagine Video 1.5 Preview through Kie API.
Availability can change when a provider retires a model, changes regional access, or makes a model temporarily unavailable. FacelessPilot may replace a model when needed, and this page will be updated when the production model set changes.
Safety controls
Image and video prompts are screened by Waffo Pancake Prompt Sift before generation. Requests that violate our Acceptable Use Policy, receive a block or review verdict, or fail provider safety checks are not generated. Third-party model providers may also apply their own input and output safety filters.
Questions about model use can be sent to [email protected].