Provider Models
The AI models Milk & Ink Studio orchestrates on your behalf. Each row tells you the model's training posture for the inputs we send it, where it's routed, and the provider's stance on output rights. This page is referenced by our Terms of Service (Section 5, Third-party AI providers) and our AI Rights & Content Policy. It covers model providers only — for the full list of every third party that processes customer data on our behalf, including hosting, database, and durable storage, see our subprocessor list.
We use the provider's no-training endpoint or the provider does not train on customer inputs by default. Safest posture for IP-sensitive work.
Inputs may be retained by the provider per their default policy. Acceptable for general use; review the linked terms before relying commercially.
Training only happens when explicitly opted in (e.g., training a custom LoRA on your own reference images at your instruction).
| Provider Model | Category | Routed via | Training posture | Output rights | Provider terms |
|---|---|---|---|---|---|
Seedance 2.0 / 2.5 Image-to-video and reference-to-video with native audio — the video model for every anime project | Video | BytePlus ModelArk (ByteDance, direct API) — automatic fal.ai fallback route to the same model during an outage | No training | Customer owns output (per ByteDance commercial license; fal.ai terms on the fallback route) | Link |
Runway Gen-4 Turbo / Gen-4.5 Image-to-video, single first-frame conditioning, 2-10s clips — the video model for every photoreal project | Video | Runway (direct developer API) | No training | Customer owns output (Runway claims no ownership of inputs or outputs and does not restrict commercial use) | Link |
Nano Banana 2 Character-identity locking (image-side) | Image | fal.ai | No training | Customer owns output | Link |
Flux LoRA Training Custom character LoRAs trained on customer's reference images | Image | fal.ai (Black Forest Labs) | No training | Customer owns the trained LoRA artifact — workspace-scoped, never used outside the customer's workspace | Link |
ElevenLabs (TTS) Eleven v3 — emotive voice synthesis with audio-tag interpretation | Audio | ElevenLabs (direct API) | No training | Customer owns generated audio (per ElevenLabs commercial license) | Link |
ElevenLabs (Music + Video-to-music) Music compose, composition_plan, video-to-music re-score | Audio | ElevenLabs (direct API) | No training | Customer owns generated audio | Link |
ElevenLabs (SFX) + fal MMAudio Prompt-only sound effects (ElevenLabs); video-aware Foley (MMAudio) | Audio | ElevenLabs / fal.ai | No training | Customer owns generated audio | Link |
fal Stable Audio Stable Audio Open — 47s music clips (fallback when ElevenLabs unconfigured) | Audio | fal.ai | No training | Customer owns generated audio (Stable Audio Open license) | Link |
Sync.so Lipsync — aligns generated audio to generated video | Audio | Sync.so (direct API) | No training | Customer owns the lipsynced output | Link |
Groq llama-3.3-70b — script planner, suggestions, copilot (default) | LLM | Groq (direct API) | No training | Customer owns generated text | Link |
Google Gemini (AI Studio) Vision + LLM — continuity scoring, multimodal review (default) | LLM | Google AI Studio (direct API) | No training | Customer owns generated assessments | Link |
What we don't promise about Provider Models
- No copyright guarantee. The U.S. Copyright Office and many other jurisdictions limit copyright protection for content generated solely by AI without sufficient human authorship. We make no representation about whether any Output is copyrightable. For commercial registration, document the human creative choices that went into the work (canon decisions, prompts, edits, selections, post-production).
- Providers can change. We may add, remove, or substitute Provider Models without notice if a provider deprecates a model or changes its terms. We surface the current set in this table and in the provider picker inside the Studio composer.
- Enterprise BYOK. Enterprise customers can route generation through their own provider keys via signed delegated-key receipts. Talk to enterprise@milkink.studio.