Ltx 2 19b Text To Video LoRA API Documentation

Ltx 2 19b Text To Video LoRA API Documentation

Playground

LTX-2 19b Text-to-Video LoRA is the first DiT-based audio-video foundation model with synchronized audio and video generation. This LoRA version supports custom style adapters for personalized video generation. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

LTX-2 Text-to-Video LoRA generates customized videos from text prompts with synchronized audio and full LoRA support. Apply up to 3 custom LoRA adapters for style personalization — perfect for branded content, custom characters, and unique visual styles.

Looking for the standard version? Try LTX-2 19B Text-to-Video without LoRA support.

Why Choose This?

• LoRA support Apply up to 3 custom LoRA adapters to personalize style, characters, or visual aesthetics.

• Synchronized audio-video generation Outputs video with matching audio in a single pass — footsteps, ambient sounds, and environmental audio fit the visual content.

• High-fidelity visuals Leverages a 19B-parameter DiT architecture for detailed, temporally consistent video.

• Flexible resolution and aspect ratio Supports 480p, 720p, and 1080p in both 16:9 (landscape) and 9:16 (vertical) formats.

• Variable duration Generate clips from 5 to 20 seconds.

Parameters

LoRA Format

Each LoRA in the loras array has:

• path (required) — URL to the LoRA weights file
• scale (optional) — Weight multiplier from 0-4, default 1

How to Use

1. Write your prompt — describe scene, action, style, and audio cues.
2. Add LoRAs — click ”+ Add Item” to include custom LoRA adapters.
3. Adjust LoRA scale — use 0.5-1.0 for subtle effects, 1.0-2.0 for stronger influence.
4. Set resolution and aspect ratio — match your target platform.
5. Set duration — start short (5s) to test, increase for final renders.
6. Run — submit and download the generated video.

Pricing

Billing Rules

• Base price: $0.10 (720p, 5 seconds)
• Resolution multiplier: 480p = 0.75×, 720p = 1×, 1080p = 1.5×
• Duration: Scales linearly (per 5 seconds)
• Total cost = duration × $0.10 × resolution_multiplier / 5

Best Use Cases

• Custom Character Animation — Apply character LoRAs for consistent identity across videos.
• Brand Content — Use trained LoRAs for consistent brand visual identity.
• Artistic Styles — Apply style LoRAs for anime, painterly, or other aesthetics.
• Social Media Content — Create unique vertical (9:16) content with custom styles.

Pro Tips

• Start with scale 1.0 and adjust up or down based on results.
• Combine LoRAs carefully — multiple LoRAs can conflict; test combinations.
• Match LoRA to content — use character LoRAs for character content, style LoRAs for aesthetics.
• Audio adapts automatically — the model generates appropriate audio even with custom styles.
• If your LoRA uses trigger words, include them in your prompt.

Notes

• Maximum video duration is 20 seconds.
• Up to 3 LoRAs can be applied simultaneously.
• LoRA version pricing is 25% higher than standard version.

Related Models

• LTX-2 19B Text-to-Video — Standard version without LoRA support.
• LTX-2 19B Image-to-Video LoRA — Image-to-video with LoRA support.
• LTX-2 19B Image-to-Video — Animate images without LoRA.

Reference

• How to use LoRA?

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",
  "resolution": "720p",
  "aspect_ratio": "16:9",
  "duration": 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/ltx-2-19b/text-to-video-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