LTX-2 AI

{ "en": "LTX-2 field guide", "zh": "LTX-2 field guide", "ru": "LTX-2 field guide", "de": "LTX-2 field guide", "fr": "LTX-2 field guide", "ja": "LTX-2 field guide", "ko": "LTX-2 field guide", "pt": "LTX-2 field guide", "tr": "LTX-2 field guide", "es": "LTX-2 field guide" }

{ "en": "LTX-2 Prompts: Practical Templates for Text-to-Video and Image-to-Video", "zh": "LTX-2 提示词:文生视频与图生视频实用模板", "ru": "LTX-2 Prompts: Practical Templates for Text-to-Video and Image-to-Video", "de": "LTX-2 Prompts: Practical Templates for Text-to-Video and Image-to-Video", "fr": "LTX-2 Prompts: Practical Templates for Text-to-Video and Image-to-Video", "ja": "LTX-2 Prompts: Practical Templates for Text-to-Video and Image-to-Video", "ko": "LTX-2 Prompts: Practical Templates for Text-to-Video and Image-to-Video", "pt": "LTX-2 Prompts: Practical Templates for Text-to-Video and Image-to-Video", "tr": "LTX-2 Prompts: Practical Templates for Text-to-Video and Image-to-Video", "es": "LTX-2 Prompts: Practical Templates for Text-to-Video and Image-to-Video" }

{ "en": "LTX-2 prompts work best when they describe one shot clearly. Put the subject and action first, then camera, environment, lighting and timing. Add constraints only when they change the shot.", "zh": "LTX-2 prompts work best when they describe one shot clearly. Put the subject and action first, then camera, environment, lighting and timing. Add constraints only when they change the shot.", "ru": "LTX-2 prompts work best when they describe one shot clearly. Put the subject and action first, then camera, environment, lighting and timing. Add constraints only when they change the shot.", "de": "LTX-2 prompts work best when they describe one shot clearly. Put the subject and action first, then camera, environment, lighting and timing. Add constraints only when they change the shot.", "fr": "LTX-2 prompts work best when they describe one shot clearly. Put the subject and action first, then camera, environment, lighting and timing. Add constraints only when they change the shot.", "ja": "LTX-2 prompts work best when they describe one shot clearly. Put the subject and action first, then camera, environment, lighting and timing. Add constraints only when they change the shot.", "ko": "LTX-2 prompts work best when they describe one shot clearly. Put the subject and action first, then camera, environment, lighting and timing. Add constraints only when they change the shot.", "pt": "LTX-2 prompts work best when they describe one shot clearly. Put the subject and action first, then camera, environment, lighting and timing. Add constraints only when they change the shot.", "tr": "LTX-2 prompts work best when they describe one shot clearly. Put the subject and action first, then camera, environment, lighting and timing. Add constraints only when they change the shot.", "es": "LTX-2 prompts work best when they describe one shot clearly. Put the subject and action first, then camera, environment, lighting and timing. Add constraints only when they change the shot." }

Hosted LTX-2.3 duration
5–20 seconds
Hosted LTX-2.3 resolutions
480p, 720p, 1080p
Modes
Text-to-video and image-to-video
Audio
Describe visual timing; add sound in a separate workflow

A dependable prompt order

Subject → action → camera → environment → lighting → mood. This order makes it easier to revise one variable without rewriting the whole prompt.

  • Use concrete verbs: walks, turns, lifts, pans, tracks.
  • Name one primary camera movement per shot.
  • Avoid stacking several unrelated scenes into one 5–20 second clip.

Text-to-video template

“[subject] [action] in [environment], [camera movement], [lighting], [mood], one continuous shot.” Replace each bracket with observable details rather than abstract adjectives.

Image-to-video template

“Preserve the composition and identity of the reference image. [subject] [small motion], [camera movement], [lighting change], stable background.” Start with subtle motion before asking for a complex transformation.

Worked prompt

A ceramic robot waters three balcony plants at sunrise, gentle left-to-right dolly, soft rim light, shallow depth of field, one continuous shot.

LTX-2.3 · text-to-video · 720p · 10 seconds · 16:9 · revise camera or action one variable at a time

The prompt is a reproducible example, not a guarantee of identical output across providers or seeds.

İş akışını çevrimiçi deneyin

Yerel kurulumu ayarlamadan önce istemi doğrulayın.

LTX-2 deneyin