Wan 2.2 i2v 720P is an ultra-fast Image-to-Video model that generates unlimited AI videos and supports custom LoRAs for personalized outputs. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Idle
$0.15per run·~66 / $10
A street musician plays a violin under a dripping canopy as rain pours around him. Wet pavement reflects neon signs. His soaked clothes cling to him as he sways to the melody. Close-up on fingers and expression. Emotional, rain-soaked urban realism.
A man warms his hands by a small campfire in the middle of a dense forest. His face glows with orange light, surrounded by darkness and faint stars above. Sparks fly upward as wood crackles. Slow zoom, introspective silence, immersive natural realism.
An elderly couple walks slowly down a tree-lined path, holding hands. Autumn leaves crunch beneath their feet. They occasionally glance at each other and smile. The camera follows from behind, capturing intimacy, warmth, and quiet companionship in soft golden light.
A teenage boy dribbles a basketball on an outdoor court at dusk. He shoots repeatedly, missing, trying again. Sweat glistens on his face.
A surfer rides a massive ocean wave, crouched low with perfect balance. Water splashes upward around him, sunlight beams through translucent blue waves. A drone-style camera circles from above, slow motion captures mid-turn splash, droplets in midair, and then returns to real-time as he finishes the ride.
A biker rides down an empty highway at midnight, city lights glowing faintly in the distance, reflections sliding across the black leather jacket. The rear camera tracks close behind, chrome parts reflecting the road, exhaust smoke trailing. Wind particles fly past, headlights flare into the camera lens as the bike leans into turns.
A scuba diver swims through an underwater cave, flashlight illuminating coral and fish swimming by. The camera trails behind the diver, bubbles rising toward the surface. Beams of light filter through holes in the rocks, floating particles create a sense of depth, occasional close-ups of marine life.
Steadicam shot plunging into dark ocean depths, bioluminescent creatures swirling around. Camera follows a diver in tactical gear, bubbles trailing from their mask. They attach a magnetic charge to a massive submarine hull, red lights blinking urgently. Suddenly, enemy divers emerge from the shadows, harpoons firing. Murky beams of light cut through the water, illuminating a frantic close-combat sequence. Final beat: the charge detonates, a shockwave of light and debris rippling through the deep — cinematic, 8K detail, James Bond × The Abyss tone.
Tracking shot along a futuristic bullet train racing through snow-covered mountains. Camera pushes in through a shattered window to reveal the hero battling masked guards in a narrow corridor, sparks flying from clashing blades. The train tilts dangerously on a curve, sending bodies sliding. In one smooth motion, the hero grapples to the roof, snow blasting against their goggles. Final beat: they leap to a passing cargo drone as the train explodes in a cascade of fire — hyper-realistic, 4K, Mission Impossible × Snowpiercer tone.
Opening with a cinematic overhead shot plunging into deep blue waters, following a diver in stealth gear. Beams of sunlight pierce the surface above, illuminating a sunken military submarine. The camera glides inside, revealing an intense underwater gunfight, bubbles exploding with every shot. The hero swims through narrow corridors, avoiding mines and searchlights. Climax: a massive torpedo detonates, sending a shockwave of debris and light, the diver propelling toward the surface as the wreck explodes beneath them.
Ultra-wide aerial shot over a convoy of armored trucks barreling through a scorching desert. The camera drops into a low ground-level tracking shot, sand blasting across the lens as a motorbike swerves between vehicles at impossible speed. RPG explosions bloom in the distance, debris raining down. The hero vaults onto a truck roof, tearing open a hatch and engaging in brutal close-quarters combat. Slow-motion beat: a grenade rolls toward the edge — the hero kicks it into the air just as it detonates, the blast silhouetting them against the orange sun.
First-person POV sprint through a neon-soaked underground data vault. Security drones zip past, their red targeting lasers slicing through clouds of smoke. The thief slides under laser grids, vaults over electric barricades, and rips open a glowing server core. The camera snap-zooms to their visor’s HUD, reflecting streams of stolen data. Final beat: an explosion behind them as they leap onto a speeding hover-bike, sparks flying, disappearing into the futuristic city skyline.
Hero fighting two masked enemies in dark alley, dynamic punches and kicks, rain reflections, dramatic lighting, photorealistic skin and clothing detail
Generate customized 720p videos from images with LoRA support using Wan 2.2 Ultra Fast. This powerful model allows you to apply custom LoRA adapters for unique styles, characters, and effects — with three different noise-level options for precise control over how LoRAs influence your output.
| Parameter | Required | Description |
|---|---|---|
| image | Yes | Source/starting image to animate (upload or public URL). |
| prompt | Yes | Text description of the motion and action you want. |
| negative_prompt | No | Elements to avoid in the generated video. |
| last_image | No | Optional ending frame for start-to-end interpolation (upload or URL). |
| duration | No | Video length: 5 or 8 seconds. Default: 5. |
| loras | No | Standard LoRA adapters to apply (up to 3). |
| high_noise_loras | No | LoRAs applied during high-noise denoising stages (up to 3). |
| low_noise_loras | No | LoRAs applied during low-noise denoising stages (up to 3). |
| seed | No | Random seed for reproducibility. Use -1 for random. |
| Enable Safety Checker | No | Toggle content safety filtering. |
Per 5-second billing based on duration.
| Duration | Calculation | Cost |
|---|---|---|
| 5 seconds | 5 ÷ 5 × $0.15 | $0.15 |
| 8 seconds | 8 ÷ 5 × $0.15 | $0.24 |
| Videos | Duration | Total Cost |
|---|---|---|
| 10 | 5s | $1.50 |
| 10 | 8s | $2.40 |
| 50 | 5s | $7.50 |
| 50 | 8s | $12.00 |
This model provides three different LoRA slots that affect different stages of the generation process:
| LoRA Type | When Applied | Best For | Max Count |
|---|---|---|---|
| loras | Throughout generation | General style, character consistency | 3 |
| high_noise_loras | Early denoising (high noise) | Overall composition, major style elements | 3 |
| low_noise_loras | Late denoising (low noise) | Fine details, textures, finishing touches | 3 |
loras for consistent style throughout.For detailed guides on using and training custom LoRAs:
Grab a WaveSpeedAI API key, then call POST https://api.wavespeed.ai/api/v3/wavespeed-ai/wan-2.2/i2v-720p-lora-ultra-fast with your input as JSON. The endpoint returns a prediction id; poll the prediction endpoint until status flips to completed, then read the output URL from data.outputs[0]. Examples for Wan 2.2 I2v 720p Lora Ultra Fast below.
# Submit the prediction
curl -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/wan-2.2/i2v-720p-lora-ultra-fast" \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $WAVESPEED_API_KEY" \
-d '{
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"image": "https://example.com/your-input.jpg",
"negative_prompt": "blurry, low quality, distorted",
"duration": 5,
"seed": -1
}'
# Response includes a prediction id. Poll for the result:
curl -X GET "https://api.wavespeed.ai/api/v3/predictions/{request_id}/result" \
-H "Authorization: Bearer $WAVESPEED_API_KEY"
# When status is "completed", read the output from data.outputs[0].// npm install wavespeed
const WaveSpeed = require('wavespeed');
const client = new WaveSpeed(); // reads WAVESPEED_API_KEY from env
const result = await client.run("wavespeed-ai/wan-2.2/i2v-720p-lora-ultra-fast", {
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"image": "https://example.com/your-input.jpg",
"negative_prompt": "blurry, low quality, distorted",
"duration": 5,
"seed": -1
});
console.log(result.outputs[0]); // → URL of the generated output# pip install wavespeed
import wavespeed
output = wavespeed.run(
"wavespeed-ai/wan-2.2/i2v-720p-lora-ultra-fast",
{
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"image": "https://example.com/your-input.jpg",
"negative_prompt": "blurry, low quality, distorted",
"duration": 5,
"seed": -1
}
)
print(output["outputs"][0]) # → URL of the generated outputWan 2.2 I2v 720p Lora Ultra Fast is a WaveSpeedAI model for AI inference, exposed as a REST API on WaveSpeedAI. Wan 2.2 i2v 720P is an ultra-fast Image-to-Video model that generates unlimited AI videos and supports custom LoRAs for personalized outputs. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing. You can call it programmatically or try it from the playground above.
POST your input parameters to the model's REST endpoint (shown in the API tab of this playground) with your WaveSpeedAI API key in the Authorization header. Submission returns a prediction ID; poll the prediction endpoint until status flips to "completed", then read the output URL from the result. The playground generates a ready-to-paste code sample in Python, JavaScript, or cURL for whatever inputs you've set. Full request/response shape is documented at https://wavespeed.ai/docs/docs-api/wavespeed-ai/wan-2.2-i2v-720p-lora-ultra-fast.
Wan 2.2 I2v 720p Lora Ultra Fast starts at $0.15 per run. That figure is the base price — the final charge scales with the parameters you set in the form (output size, length, count, references, or whatever knobs this model exposes), so a higher-quality or larger output costs more than a minimal one. The exact cost for your current input is shown live next to the Generate button before you submit, and the actual per-call charge is recorded on the prediction afterwards.
Key inputs: `prompt`, `image`, `duration`, `seed`, `negative_prompt`, `high_noise_loras`. The full JSON schema (types, defaults, allowed values) is rendered above the Generate button and mirrored in the API reference at https://wavespeed.ai/docs/docs-api/wavespeed-ai/wan-2.2-i2v-720p-lora-ultra-fast.
Average end-to-end generation time on WaveSpeedAI is around 80 seconds per request — measured across recent runs. Queue time scales with global demand; live status is visible in the prediction record.
Commercial usage rights depend on the model's license, set by its provider (WaveSpeedAI). The license summary appears on the model card above; see WaveSpeedAI's Terms of Service for platform-level conditions.