Some scenes need more time to unfold — a slow reveal, an emotional exchange, a panoramic build that lets your viewer linger.
That’s where long shots come in.
With LTX-2, you can generate up to 20 seconds of continuous motion — perfect for moments that breathe, evolve, and hold emotion.
But to make those seconds count, your prompting strategy needs to stretch with them.
1. Why Longer Shots Matter
Longer shots do what short cuts can’t — they let emotion build in real time.
They’re ideal for:
- Dramatic reveals (a door opening, a skyline emerging through fog)
- Panoramic storytelling (establishing a sense of world or tone)
- Emotional exchanges (conversations, hesitation, connection)
- Moments of stillness where the audience can feel time pass
Eight-second generations are great for rhythm. But when your scene needs to breathe, twenty seconds opens the door to cinematic pacing.
2. Longer Shots Need Longer Prompts
More runtime means more structure. Treat your long-shot prompt like a mini scene — not just a description.
Well, sort of. It’s not just about writing more — it’s about writing with sequence and timing in mind.
Order your prompt by actions and make sure the amount of movement fits comfortably within your selected duration. Too much detail, and the motion will feel rushed. Too little, and you’ll see unwanted pauses or drifting behaviour. Getting the pacing right is a bit of trial and error, so start simple and build from there.
A clear format helps:
- Scene header — place and time
- Short description — tone and atmosphere
- Blocking — how subjects move in sequence with the camera
- Dialogue and cues — use quotes for dialogue and (brackets) for tone or performance direction
This keeps your prompt readable and helps LTX-2 understand flow and intention.
Tips for Realism and Consistency
- Start in a close-up, then move out — it grounds the scene and helps the model retain facial and material detail.
- Wider shots can soften likeness across all AI models. To avoid this, have your character turn away or stay at a consistent distance from the lens.
- Avoid abrupt reframing or zooms — smooth, natural motion reads best over 20 seconds.
If dialogue or performance cues end too early, add a soft closing action such as character takes a beat and looks up or camera drifts upward as the wind passes to carry the motion to the finish.
3. Image-to-Video vs. Text-to-Video
Both image-to-video and text-to-video workflows handle long shots beautifully — but each has a different creative role.
Image-to-video gives you control over style, framing, and character consistency. It’s ideal when you’ve already established a look and want to preserve it across multiple shots or scenes.
Text-to-video, on the other hand, is more exploratory. It’s perfect when you’re building a new moment from scratch, experimenting with composition, or testing how actions and pacing play out over 20 seconds.
Here’s how the same kind of scene plays out in each approach:
Example 1 – Image-to-Video
EXT. EMPTY STREET – DAWN
The shot opens in an extreme close-up of a woman’s face.
She stands still, quiet, eyes steady.
The camera holds for a moment.
Then she turns and begins to walk away.
The camera tracks her from behind as she moves through the empty street.
No cars. No people. Only her footsteps fading as she disappears down the block.
The camera slows to a stop, holding on the empty street.
The sound of the wind lingers before the shot ends.
This version is suited for image-to-video, where style and likeness remain consistent as the character moves through space.
Example 2 – Text-to-Video
EXT. DAY. TOWN.
Extreme close-up of an older woman.
The camera then pulls out and the woman turns around.
She walks along an abandoned street in a town.
The camera tracks her.
This shorter, cleaner structure works well for text-to-video, where the model generates both visuals and motion from scratch. It defines pacing and key actions without locking in visual references — ideal for testing timing or building out first drafts of longer sequences.
4. Describe the Blocking
Blocking — the choreography between subject and camera — is what makes long shots feel cinematic. Describe how the action and camera move together, not just what’s visible.
Example:
EXT. EMPTY TOWN SQUARE – MORNING
The shot opens in an extreme close-up of an older woman’s face, sunlight glinting in her silver hair.
Her eyes are calm, thoughtful. The air is still.
She turns slightly, looking directly into the camera.
For a moment, she just holds the gaze — steady, composed.
WOMAN (quietly):
“Funny how quiet it gets.”
She takes a breath, glances toward the empty street, then looks back to the lens.
A small, knowing smile crosses her face.
She nods once, almost to herself.
After a beat, she turns and begins walking down the street.
Her footsteps echo softly on the stone.
The town is silent — no voices, no movement, only the faint hum of wind.
The camera follows her for a few steps, then slows as she moves away.
She turns a corner and disappears, leaving the frame empty.
For a moment, the shot holds on the quiet square.
Then, slowly, the camera begins to pan right.
The soft rumble of an old tractor grows in the distance.
The tractor rolls gently into view, its faded paint catching the morning light.
At the wheel sits a man wearing a flat cap, posture relaxed but steady.
He drives slowly through the square, the sound of the engine low and rhythmic.
The camera continues tracking, sliding alongside before easing closer to the cab.
The frame tightens into a close-up on the man’s face.
His hands rest firmly on the steering wheel, veins and calluses visible, the light tracing every detail.
He looks ahead, quiet and focused, as the tractor hums beneath him.
The camera holds — a still moment of motion — before the shot ends.
5. Direct the Performance
Dialogue and expression give your shot emotional rhythm. Even without sound, lines of dialogue or bracketed cues can shape pacing and movement.
Example:
EXT. EMPTY TOWN SQUARE – MORNING
The shot opens in an extreme close-up of an older woman’s face, sunlight glinting in her silver hair.
Her eyes are calm, thoughtful. The air is still.
She turns slightly, looking directly into the camera.
For a moment, she just holds the gaze — steady, composed.
WOMAN (quietly):
“Funny how quiet it gets.”
She takes a breath, glances toward the empty street, then looks back to the lens.
A small, knowing smile crosses her face.
She nods once, almost to herself.
After a beat, she turns and begins walking down the street.
Her footsteps echo softly on the stone.
The town is silent — no voices, no movement, only the faint hum of wind.
The camera follows her for a few steps, then slows as she moves away.
She turns a corner and disappears, leaving the frame empty.
For a moment, the shot holds on the quiet square.
Then, slowly, the camera begins to pan right.
The soft rumble of an old tractor grows in the distance.
The tractor rolls gently into view, its faded paint catching the morning light.
At the wheel sits a man wearing a flat cap, his posture relaxed but steady.
He drives slowly through the square, the sound of the engine low and rhythmic.
The camera continues tracking, sliding alongside before easing closer to the cab.
The frame tightens into a close-up on the man’s face.
His hands rest firmly on the steering wheel, veins and calluses visible, the morning light tracing every detail.
He looks ahead, quiet and focused, as the tractor hums beneath him.
The camera holds — a still moment of motion — before the shot ends.
6. Multi-Character Scenes
When two characters share a scene, use pacing and camera focus to manage the exchange. Let the camera linger on one speaker before moving to the next. Avoid abrupt cuts; instead, flow naturally through reactions and pauses.
Example:
EXT. EMPTY TOWN SQUARE – MORNING
The shot opens in an extreme close-up of an older woman’s face, sunlight glinting in her silver hair.
Her eyes are calm, thoughtful. The air is still.
She turns slightly, looking directly into the camera.
For a moment, she just holds the gaze — steady, composed.
WOMAN (quietly):
“Funny how quiet it gets.”
She takes a breath, glances toward the empty street, then looks back to the lens.
A small, knowing smile crosses her face.
She nods once, almost to herself.
After a beat, she turns and begins walking down the street.
Her footsteps echo softly on the stone.
The town is silent — no voices, no movement, only the faint hum of wind.
The camera follows her for a few steps, then slows as she moves away.
She turns a corner and disappears, leaving the frame empty.
For a moment, the shot holds on the quiet square.
Then, slowly, the camera begins to pan right.
The soft rumble of an old tractor grows in the distance.
The tractor rolls gently into view, its faded paint catching the morning light.
At the wheel sits a man wearing a flat cap, posture relaxed but steady.
He drives slowly through the square, the sound of the engine low and rhythmic.
The camera continues tracking, sliding alongside before easing closer to the cab.
The frame tightens into a close-up on the man’s face.
His hands rest firmly on the steering wheel, veins and calluses visible, the morning light tracing every detail.
He glances toward the direction the woman walked, then looks ahead again.
A small smile flickers.
MAN (quietly, to himself):
“Still is.”
The camera holds on him as the tractor rolls on, the hum of the engine fading softly into the stillness.
7. Final Thoughts
Long shots in LTX-2 aren’t just about duration. They’re about momentum, intention, and rhythm.
Use those 20 seconds to guide the audience through a living moment — every beat, every movement, every breath. Every line in your prompt is a cue to the camera. Every bracketed note is a director’s whisper. Build your scene like you’re on set — and let LTX-2 bring it to life.










.png)
