Wan 2.1 I2V 480p LoRA
Playground
Try it on WavespeedAI!Generate unlimited 480P AI videos with WAN 2.1 Image-to-Video and custom LoRA support for personalized styles. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Features
WAN 2.1 Image-to-Video 480p LoRA — wavespeed-ai/wan-2.1/i2v-480p-lora
wavespeed-ai/wan-2.1/i2v-480p-lora turns a single input image into a short, coherent video clip at 480p, with optional LoRA effects for stylized motion, transformations, or consistent visual “behavior” across runs.
What it’s good at
- Image-to-video animation from one still image (great for product shots, portraits, scenes)
- LoRA-driven motion/effects (for example: rotate, zoom-call framing, playful VFX-style transformations)
- Fast iteration for storyboard-style testing at a lightweight resolution
Inputs
-
Prompt (required): describe the action, camera, and mood
-
Image (required): the starting frame
-
Negative prompt (optional): reduce artifacts or unwanted styles
-
LoRAs (optional): add one or more LoRAs with:
- Path: LoRA identifier
- Scale: LoRA strength
-
Size: 480p
-
Duration: 5s or 10s
-
Num inference steps: higher can improve detail/coherence (slower)
-
Guidance scale: higher follows the prompt more strictly
-
Flow shift: controls motion feel (lower is calmer, higher is more dynamic)
-
Seed: set for repeatable results
How to use (Playground)
- Upload your image.
- Write a prompt that includes subject + action + camera movement + environment.
- (Optional) Add a LoRA and start with a moderate scale, then adjust upward only if the effect is too subtle.
- Keep guidance scale moderate for natural motion; increase it if the model ignores your requested action.
- Run at 5s first, then move to 10s once the motion looks right.
Prompting tips (I2V)
- Start with a single clear action: walks, turns, smiles, waves, looks back, picks up an object.
- Add camera direction in plain film language: slow push-in, handheld, dolly left, tilt up, rack focus.
- If you use an “effect” LoRA (like deflate/rotate), describe the target as a non-graphic, clearly fictional or toy-like transformation to avoid unwanted safety filtering.
Example prompt style (safe wording): A beach photo becomes a playful VFX shot: the character turns into an inflatable toy version of itself and slowly deflates like a balloon, then settles gently on the sand. Smooth camera push-in, soft sunlight, natural motion, no distortion.
Use cases
- Social content and short ads: animate a key visual into a 5–10s loopable clip
- Product hero animation: subtle parallax, gentle rotation, light camera push for ecommerce
- Stylized LoRA effects: rotate, “video call” framing, playful toon-like transformations
- Prototype storytelling: turn concept art into quick motion beats for pitches
Pricing
| Model | Resolution | Duration | Price per run |
|---|---|---|---|
| wavespeed-ai/wan-2.1/i2v-480p-lora | 832×480 | 5s | $0.20 |
| wavespeed-ai/wan-2.1/i2v-480p-lora | 832×480 | 10s | $0.30 |
Notes
- 480p is ideal for fast iteration, previews, and high-volume generation.
- LoRA effects can overpower the base motion if scaled too high; increase gradually.
- Safety checks may block certain real-person transformations; using fictional/toy framing typically works better.
Authentication
For authentication details, please refer to the Authentication Guide.
API Endpoints
Submit Task & Query Result
# Submit the task
curl --location --request POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/wan-2.1/i2v-480p-lora" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
"loras": [
{
"path": "Remade-AI/Deflate",
"scale": 1
}
],
"size": "832*480",
"num_inference_steps": 30,
"duration": 5,
"guidance_scale": 5,
"flow_shift": 3,
"seed": -1
}'
# Get the result
curl --location --request GET "https://api.wavespeed.ai/api/v3/predictions/${requestId}/result" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}"
Parameters
Task Submission Parameters
Request Parameters
| Parameter | Type | Required | Default | Range | Description |
|---|---|---|---|---|---|
| image | string | Yes | - | The image for generating the output. | |
| prompt | string | Yes | - | The positive prompt for the generation. | |
| negative_prompt | string | No | - | The negative prompt for the generation. | |
| loras | array | No | max 3 items | List of LoRAs to apply (max 3). | |
| loras[].path | string | Yes | - | Path to the LoRA model | |
| loras[].scale | float | Yes | - | 0.0 ~ 4.0 | Scale of the LoRA model |
| size | string | No | 832*480 | 832*480, 480*832 | The size of the generated media in pixels (width*height). |
| num_inference_steps | integer | No | 30 | 1 ~ 40 | The number of inference steps to perform. |
| duration | integer | No | 5 | 5, 10 | The duration of the generated media in seconds. |
| guidance_scale | number | No | 5 | 0.00 ~ 20.00 | The guidance scale to use for the generation. |
| flow_shift | number | No | 3 | 1.0 ~ 10.0 | The shift value for the timestep schedule for flow matching. |
| seed | integer | No | -1 | -1 ~ 2147483647 | The random seed to use for the generation. -1 means a random seed will be used. |
Response Parameters
| Parameter | Type | Description |
|---|---|---|
| code | integer | HTTP status code (e.g., 200 for success) |
| message | string | Status message (e.g., “success”) |
| data.id | string | Unique identifier for the prediction, Task Id |
| data.model | string | Model ID used for the prediction |
| data.outputs | array | Array of URLs to the generated content (empty when status is not completed) |
| data.urls | object | Object containing related API endpoints |
| data.urls.get | string | URL to retrieve the prediction result |
| data.has_nsfw_contents | array | Array of boolean values indicating NSFW detection for each output |
| data.status | string | Status of the task: created, processing, completed, or failed |
| data.created_at | string | ISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”) |
| data.error | string | Error message (empty if no error occurred) |
| data.timings | object | Object containing timing details |
| data.timings.inference | integer | Inference time in milliseconds |
Result Request Parameters
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
| id | string | Yes | - | Task ID |
Result Response Parameters
| Parameter | Type | Description |
|---|---|---|
| code | integer | HTTP status code (e.g., 200 for success) |
| message | string | Status message (e.g., “success”) |
| data | object | The prediction data object containing all details |
| data.id | string | Unique identifier for the prediction, the ID of the prediction to get |
| data.model | string | Model ID used for the prediction |
| data.outputs | string | Array of URLs to the generated content (empty when status is not completed). |
| data.urls | object | Object containing related API endpoints |
| data.urls.get | string | URL to retrieve the prediction result |
| data.status | string | Status of the task: created, processing, completed, or failed |
| data.created_at | string | ISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”) |
| data.error | string | Error message (empty if no error occurred) |
| data.timings | object | Object containing timing details |
| data.timings.inference | integer | Inference time in milliseconds |