Wan 2.1 T2V 480p Ultra Fast

Wan 2.1 T2V 480p Ultra Fast

Playground

WAN 2.1 T2V 480p Ultra-Fast turns text prompts into unlimited 480p AI videos with ultra-fast throughput and reliable 480p output. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Wan 2.1 T2V 480p Ultra Fast — wavespeed-ai/wan-2.1/t2v-480p-ultra-fast

Wan 2.1 T2V 480p Ultra Fast is a text-to-video model optimized for speed and iteration. Write a prompt, pick a duration, and generate a short 480p video quickly—ideal for rapid concept testing, storyboard previews, and high-volume variations.

What it’s good at

• Text-to-Video (T2V): generate video directly from a prompt
• Ultra-fast generation: built for quick iteration and high-throughput workflows
• 480p output: lightweight resolution for draft-quality previews and batch runs
• Creative control knobs: guidance scale, steps, flow shift, seed

Common use cases

• Storyboard & pre-visualization: turn scripts into short motion sketches before committing to higher-cost renders
• Marketing concept exploration: generate multiple variants of the same idea (camera, action, mood) and pick winners fast
• Motion style prototyping: test different motion intensity settings for pacing, dynamism, and scene continuity
• Batch ideation: generate many drafts to explore composition, timing, and cinematic language

Inputs

• Prompt (required): describe subject, action, setting, lighting, camera movement, and mood
• Negative prompt (optional): list artifacts to avoid (blur, jitter, distortion, low quality, inconsistent motion)

Parameters (Playground / API)

• Size: 832×480 (480p)
• Duration: 5s or 10s
• Num inference steps: higher can improve stability/detail but costs more time
• Guidance scale: higher follows the prompt more strictly; too high may reduce natural motion
• Flow shift: adjusts motion feel and temporal dynamics
• Seed: use a fixed value for reproducibility; use -1 for random

Pricing

Prompting tips (T2V)

• Write like a director: Subject + Action + Scene + Camera + Style
• Use concrete motion verbs: walks, turns, reaches, leans, spins, drifts, stops
• Add camera language to make it feel like “video,” not “animated still”: slow dolly-in, handheld, orbit, pan left, tilt up, rack focus
• Keep negative prompts short and practical: blurry, jittery, flicker, distorted, low quality, inconsistent motion

# Submit the task
curl --location --request POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/wan-2.1/t2v-480p-ultra-fast" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "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}"