
Wan 2.2 Turbo Spicy Infinite Image-to-Video LoRA API by Atlas Cloud
Image-to-video LoRA variant for segmented prompt video generation with stable motion and 30fps workflow post-processing.
Wan 2.2 Turbo Spicy Infinite Image-to-Video — LoRA
Model Overview
| Field | Description |
|---|---|
| Model Name | atlascloud/wan-2.2-turbo-spicy/infinite-image-to-video-lora |
| Model Type | Advanced Image-to-Video Generation |
| Core Architecture | Mixture-of-Experts (MoE) |
| Active Parameters | 14B + LoRA adapter |
| Variant | LoRA |
| Tuning | Spicy-tuned post-processing pipeline (adult-oriented) |
The LoRA variant of Wan 2.2 Turbo Spicy Infinite Image-to-Video. Same Infinite segmented-prompt mechanic and acceleration stack as the base model, with LoRA-grade fidelity and motion stability for final renders. Built on the Wan 2.2 Mixture-of-Experts (MoE) foundation with a spicy-tuned post-processing pipeline for adult-oriented creative work.
Key Features & Innovations
1. Ultra-Fast Inference: 4-Step Distillation with RCM
- RCM (Refined Consistency Model) Sampler — efficient ODE solver that improves single-step sampling quality.
- 4-Step Distillation — denoising compressed to 4 steps, enabling cinematic-grade generation at low latency. LoRA inference is ~10–20 % slower than base but stays well within interactive territory.
2. Infinite-Length Generation: Anchor-Frame Autoregressive Architecture
- Anchor-Frame Evolution — automatically extracts key "anchor frames" during generation as global temporal references.
- Dual-Frame Constraint (Anchor + Last Frame) — combines global structural consistency with motion continuity to construct video sequences autoregressively.
- Semantic Stability — LoRA further sharpens identity and detail consistency across multi-minute outputs.
3. Cinematic-Level Aesthetics (Inherited + LoRA-Enhanced)
- Precise Control — detailed labels for lighting, composition, color tone.
- Complex Motion — fluid motion across diverse semantics.
- Fine-Grained Fidelity — LoRA adapter delivers sharper textures, more stable identities, and stylistic depth that the base variant cannot match on its own.
Why Infinite?
Output duration equals prompt_count × duration_per_segment, up to 6 prompts x 5 s. Direct each segment with its own prompt; the API returns one server-stitched 30 fps MP4.
| Prompts | Per-segment | Total output |
|---|---|---|
| 1 | 5 s | 5 s |
| 3 | 5 s | 15 s |
| 6 | 5 s | 30 s |
When to Pick the LoRA Variant
- Final renders, not drafts — the quality margin is worth the +30 % price.
- Subjects with fine identity details that must stay consistent across segments.
- Stylized motion or lighting that the base model under-delivers on.
For early iteration / bulk drafts, use the base:
atlascloud/wan-2.2-turbo-spicy/infinite-image-to-video (cheaper, faster).
60-second Quickstart
curl -X POST https://api.atlascloud.ai/api/v1/model/generateVideo \ -H "Authorization: Bearer $APIKEY" \ -H "Content-Type: application/json" \ -d '{ "model": "atlascloud/wan-2.2-turbo-spicy/infinite-image-to-video-lora", "image": "https://static.atlascloud.ai/media/images/db548fe3bd5cafa4ef7e0141d69c8566.jpeg", "prompt": [ "She turns slowly toward the camera, golden hour light hitting her face.", "She walks forward through the wheat field, hand brushing the tops.", "Close-up: a single tear catches the sun as she smiles." ], "duration": 5, "resolution": "720p" }'
Returns one MP4 — segments are stitched server-side at 30 fps.
Request Fields
| Field | Type | Required | Notes |
|---|---|---|---|
model | string | ✅ | atlascloud/wan-2.2-turbo-spicy/infinite-image-to-video-lora |
image | string (URL) | ✅ | Source frame; jpg/png |
prompt | string[] | ✅ | Must be a JSON array. Plain string is rejected. |
duration | number | ✅ | Fixed at 5 s per segment. |
resolution | string | optional | 480p, 720p, or 1080p. Defaults to 720p. |
seed | number | optional | -1 for random |
Pricing — at a glance
price = $0.026 × max(1, prompt_count) × max(5, duration_seconds) × resolution_factor 480p → 1 720p → 2 1080p → 3
Common combos:
| Prompts | Duration | Resolution | Total |
|---|---|---|---|
| 1 | 5 s | 480 p | $0.13 |
| 1 | 5 s | 720 p | $0.26 |
| 1 | 5 s | 1080 p | $0.39 |
| 3 | 5 s | 720 p | $0.78 |
| 6 | 5 s | 720 p | $1.56 |
| 6 | 5 s | 1080 p | $2.34 |
Output Spec
- Format: MP4 (H.264)
- Frame rate: 30 fps (post-processed)
- Resolution: 480 p / 720 p / 1080 p tiers, aspect-ratio preserving
- Audio: none
Intended Use & Applications
- Final cinematic renders with cross-segment identity stability.
- High-fidelity advertising / pre-visualization that depend on stylistic consistency.
- Identity-critical I2V where minor drift would break the narrative.
Content Policy
This model is tuned for adult-oriented (NSFW) generation. By calling it you confirm:
- All depicted subjects are 18 +.
- You hold the rights to the source image.
- You will not generate content depicting real, identifiable people without their explicit consent.
Violations may result in account suspension.
Limitations
promptmust be a JSON array, never a plain string.- LoRA reduces but does not eliminate cross-segment identity drift.
- LoRA generation is ~10–20 % slower per segment than base.
Related
- Base variant:
atlascloud/wan-2.2-turbo-spicy/infinite-image-to-video - Non-spicy LoRA alias:
atlascloud/wan-2.2-turbo/infinite-image-to-video-lora
Note: This model is designed to empower the creative community. Users are expected to follow AI ethical guidelines and copyright regulations.


















