Seedance 2.5 Spicy ref2v: NSFW Video Without a Reshoot

Re-describing the same outfit, face, and room from scratch every time a scene changes is why most NSFW AI video sets drift: skin tone shifts, the robe becomes a different robe, the room gets a different lamp. bytedance/seedance-2.5-ref2v-spicy skips that problem by generating from reference material instead of a fresh text description alone — feed in the images, video, or audio that already define the subject, and the new generation carries them forward without a reshoot.

Key Takeaways

• ref2v generates from reference inputs — images, video, audio — not a blind text or single-image prompt; the subject, outfit, and scene carry over without new reference material.
• Reference limits (docs.siray.ai OpenAPI spec, verified today): up to 30 images, 10 videos, and 10 audio clips per generation.
• size runs 480p / 720p / 1080p, duration 4-30 seconds (or -1 for auto), and aspect_ratio spans 7 options.
• Console pricing: 480p $0.140/s, 720p $0.302/s, 1080p $0.739/s, no discount currently applied.

How ref2v Differs From t2v and i2v

t2v-spicy starts a shot from text alone. i2v-spicy starts from one image and carries it into a shot. ref2v-spicy goes further: it accepts arrays of reference images, reference videos, and reference audio at once, and generates a new video that keeps what those references define — face, outfit, set — while the prompt only has to handle what changes, like a new pose or camera move. That's a different mechanism from our single continuous multi-beat shot approach, which uses t2v-spicy/i2v-spicy to block 2-3 beats inside one 30-second take. This is about consistency across separate generations — a new scene next week that still matches last week's shoot, without a new reference photo session each time. If prompt wording alone is enough for your case, our character consistency guide covers that route on t2v/i2v; ref2v is the option once wording stops being enough.

A Directed ref2v Prompt

model: bytedance/seedance-2.5-ref2v-spicy
duration: 10
size: "1080p"
aspect_ratio: "9:16"
audio_enable: true
images: ["ref-face-01.jpg", "ref-outfit-front.jpg", "ref-outfit-back.jpg"]
prompt: "The woman from the reference images now stands at a hotel window atnight, city lights behind her, wearing the same red robe from the references.She turns slowly toward camera, robe slipping off one shoulder, and settlesinto a seated pose on the windowsill. Warm practical lighting, same skin toneand hair as the references, continuous handheld-style motion."

The reference images anchor face, hair, and wardrobe, so the prompt only has to describe the new setting and movement. Swap in a reference video instead of stills when the point is to carry a specific motion or expression forward, not just an appearance.

What It Costs

The console model-detail page's Pricing block is the real number to use; the public pricing page only lists the 480p floor. For the full spec and price comparison against Wan 3.0 Spicy's own ref2v endpoint, see our side-by-side breakdown.

FAQ

How is ref2v different from i2v on Seedance 2.5 Spicy?

i2v-spicy starts a shot from a single image. ref2v-spicy accepts up to 30 reference images plus reference video and audio, so more of what defines the subject carries into the new generation at once.

How many reference images can I actually use?

Up to 30 images, 10 videos, and 10 audio clips per generation, confirmed against the live OpenAPI schema on docs.siray.ai.

Is the 30-second duration unique to this model?

No — Wan 3.0 Spicy also runs up to 30 seconds. ref2v-spicy's real advantage is the reference inputs, not an exclusive duration ceiling.

Create your free Siray account and generate your next scene from what you already shot, not a new session.

Compliance: Siray is committed to responsible AI use. All content generated through the platform must comply with applicable laws. Siray maintains zero tolerance for CSAM and any illegal content. Uncensored generation capabilities are provided exclusively for lawful adult content creation by and for consenting adults aged 18 and over.