All 11 workflow variants | Landscape keyframes 1344x768 | Custom speech audio 8.9s
gemma-3-12b-it-Q2_K.gguf), or offload text encoding to an API. Enable tiled VAE decode and the VRAM management node to fit within memory.flashvsr_v2v_upscaleupscale_v2vframe_interpolation_v2v
First Frame (facing camera)

Middle Frame (head turned right)

Last Frame (facing camera, warm smile)
Text-to-Video with two-pass spatial upscaling and generated audio.
Make this image come alive with fluid motion. A beautiful blonde woman smiles towards the viewer. Then she talks and says with a soft female british voice : "Lets go and make some great videos with LTX 2.3". Steady camera slowly panning in towards the main subject. The scene is set to a sunny windy day outside with the city in the background. Ambient low volume city sound in the background.
Text-to-Video via vMonad MediaGen ltx23_t2v strategy (ComfyUI workflow template).
A beautiful blonde woman smiles towards the viewer. Then she talks and says with a soft female british voice: "Lets go and make some great videos with LTX 2.3". Steady camera slowly panning in towards the main subject. The scene is set to a sunny windy day outside with the city in the background. Ambient low volume city sound in the background.
Text-to-Video single-pass. Fastest generation, no upscale.
Make this image come alive with fluid motion. A beautiful blonde woman smiles towards the viewer. Then she talks and says with a soft female british voice : "Lets go and make some great videos with LTX 2.3". Steady camera slowly panning in towards the main subject. The scene is set to a sunny windy day outside with the city in the background. Ambient low volume city sound in the background.
Image-to-Video with two-pass spatial upscaling.
Make this image come alive with fluid motion. A beautiful blonde woman smiles towards the viewer. Then she talks and says with a soft female british voice : "Lets make some great videos with LTX 2.3. Sometimes the sentence needs to extended to avoid cutting into a random scene". Steady camera slowly panning in towards the main subject.
Image-to-Video single-pass. Faster, no upscale.
Make this image come alive with fluid motion. A beautiful blonde woman smiles towards the viewer. Then she talks and says with a soft female british voice : "Lets make some great videos with LTX 2.3. Sometimes the sentence needs to extended to avoid cutting into a random scene. My image compression doesn't seem to have any effect on me". Steady camera slowly panning in towards the main subject.
Image-to-Video using GGUF quantized model for lower VRAM.
Make this image come alive with fluid motion. A beautiful blonde woman smiles towards the viewer. Then she talks and says with a soft female british voice : "Lets make some great videos with LTX 2.3. This is a gguf workflow for low vram support. Steady camera slowly panning in towards the main subject.
Image-to-Video with custom audio input overlay.
A woman in a red dress on a coastal cliff at sunset. She faces the camera and speaks directly to the viewer. Her lips move as she says: "The old gods are silent. But I am not. I have walked through fire and shadow, and I am still here." Solo shot, one person. Wind in her hair. Golden hour lighting.
First + Last frame injection. Teacher walks from blackboard to window in classroom scene.
Make this image come alive with cinematic motion, smooth animation. A beautiful teacher is standing next to a blackboard. On the blackboard it says "LTX-2.3". The teacher is looking towards the viewer. She starts walking across the classroom She talks and then she says "The kids dont know it, but I will not call to class today. I am too busy playing around with LTX 2 point 3" She walks over to the window, outside there as kids playing. "They are playing around. So will I" cinematic, photo realistic. Highly detailed and intricate
First + Last frame interpolation with custom audio. Red dress to warrior armor transformation.
A young woman stands on a coastal cliff at golden hour sunset, facing the camera. Wind blows her dark wavy hair. Her red dress transforms into gleaming silver-gold warrior armor with ornate shoulder pauldrons and engraved chest plate. The transformation ripples across her body like liquid metal. Same woman, same background. Solo shot. Steady camera.
First + Middle + Last frame guider. Coca-Cola bar advertisement with bottle opening and drinking sequence.
Make this image come alive with cinematic motion, smooth animation. Cinematic high-quality Coca-Cola advertisement video. The scene starts with a close-up of a capped Coca-Cola glass bottle on a dark wood bar counter. The woman in the background says "That looks tempting." She leans forward and grabs the Coca-Cola bottle, twists off the cap and places it on the counter. She says "Oh, its one of those classic ones. Refreshing." The woman stands up, holds the open Coca-Cola bottle, and says "A Coke a day keeps the boredom away." She tilts her head back and drinks from the open bottle. Ambient bar room sounds in the background. Cinematic, photorealistic, warm amber lighting throughout
First + Middle + Last frame injection with custom audio. Coca-Cola bar ad with SRT-timed lip sync.
Cinematic Coca-Cola bar advertisement. Warm amber lighting. Same woman with dark wavy hair in black top throughout. Solo shot. [0.0s] Woman at bar counter, blurred Coca-Cola bottle in foreground. [0.0s-1.1s] She says "That looks tempting." [1.1s-3.2s] She reaches forward, grabs the Coca-Cola bottle, twists off the cap, places cap on counter. [3.2s-4.8s] She holds the open bottle and says "Mmm, refreshing." [4.8s-7.0s] She slowly raises the bottle toward her face. [7.0s-8.8s] She says "A Coke a day keeps the boredom away." [8.8s-9.8s] She tilts head back and drinks from the open bottle. Cinematic, photorealistic. Warm amber bar lighting.
FML2V Guider with custom audio. F-16 cockpit fighter pilot scene with lip-sync.
Make this image come alive with cinematic motion, smooth animation. The woman is talking, and she says: "Tower, this is Maverick. I have visual on the target. Engaging afterburners. Let's show them what we've got." Perfect lip-sync to the attached audio.