Wan 2.1 T2V 720p LoRA

Wan 2.1 T2V 720p LoRA

Playground

Wan 2.1 Text-to-Video 720P creates 720P videos from text prompts and supports custom LoRAs for personalized styles. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Generate stunning videos from text descriptions with Wan 2.1 Text-to-Video 720p LoRA. This powerful model transforms your written prompts into high-quality 720p videos with smooth motion and cinematic quality — plus full LoRA support for custom styles, characters, and aesthetics.

Why It Stands Out

• Pure text-to-video generation: No source image needed — describe your vision and watch it come to life.
• LoRA support: Load custom LoRA models to apply specific styles, maintain character consistency, or match brand aesthetics.
• Prompt-guided creation: Control scenes, actions, camera movements, and atmosphere through natural language.
• Negative prompt support: Exclude unwanted elements for cleaner, more controlled outputs.
• Flexible duration: Generate 5-second or 10-second clips depending on your needs.
• Reproducibility: Use the seed parameter to recreate exact results or iterate on variations.

Pricing

Parameters

How to Use

1. Write a prompt describing the scene, action, and style you want.
2. Add a LoRA (optional) — paste the URL to your custom LoRA and set the strength.
3. Set parameters — adjust duration, guidance scale, and other settings as needed.
4. Add a negative prompt (optional) to exclude unwanted elements.
5. Click Run and wait for your video to generate.
6. Preview and download the result.

How to Use LoRA

LoRA (Low-Rank Adaptation) lets you customize the model’s output style without retraining the full model.

• Use your LoRA: Host your .safetensors file at a public URL and paste it into the lora_url field.
• Train your LoRA: Learn how to create custom LoRAs in our guide: Train Your Own LoRA Model

Common LoRA use cases: consistent character appearance, specific art styles, brand-aligned aesthetics, anime/cartoon styles.

Best Use Cases

• Social Media Content — Create scroll-stopping video content from scratch.
• Marketing & Advertising — Produce concept videos and ad creatives without filming.
• Storytelling & Animation — Generate scenes for short films, music videos, or narrative projects.
• Game & App Development — Create promotional trailers and UI animations.
• Personalized Content — Use custom LoRAs for branded or character-consistent videos.

Pro Tips for Best Quality

• Be specific in your prompt — describe subject, action, environment, lighting, and camera movement.
• Use negative prompts to reduce common artifacts: blur, distortion, jitter, or watermarks.
• Start with lower inference steps (20–25) for quick previews, then increase for final renders.
• When using LoRA, start with strength around 0.7 and adjust based on results.
• Fix the seed when iterating to isolate the effect of parameter changes.

Notes

• Ensure any LoRA URLs are publicly accessible.
• Processing time varies based on duration and current queue load.
• Please ensure your prompts comply with content guidelines.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "size": "1280*720",
  "num_inference_steps": 30,
  "duration": 5,
  "guidance_scale": 5,
  "flow_shift": 5,
  "seed": -1
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/wan-2.1/t2v-720p-lora" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL=$(printf '%s' "${TASK}" | jq -r '.urls.get // empty')
if [ -z "${RESULT_URL}" ]; then RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"; fi

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    created|processing) sleep 2 ;;
    *) printf 'Unexpected status: %s
' "${STATUS}" >&2; exit 1 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters