wan-2-2/first-last-frame
Streamline scene design with high-fidelity, auto-interpolated video
Start from the main model page: LTX 2 Pro Text-to-Video. LTX 2 Pro Image-to-Video lets you animate images with LTX 2 into cinematic, high-fidelity clips with stable motion and optional synchronized audio - ideal when you prioritize quality over fast text-to-video generation throughput.
LTX 2 Pro Image-to-Video (I2V) turns a single reference image plus a prompt into a coherent video. It's built for maximum fidelity: preserve subject identity, add controlled camera/motion, and optionally generate synchronized audio. Use it to animate images with LTX 2 for cinematic results.
image_url.prompt describes motion, atmosphere, and camera moves (e.g., "slow dolly in", "pan right").Use I2V to create a strong first pass, then refine with targeted edits:
1) Generate your base clip here.
2) Send it to Retake to adjust motion intensity, re-time segments, or refine specific regions without starting over: LTX 2 Retake Video
image_url (required): JPG/PNG/WEBP/... reference image.prompt (required): Describe subject, motion, environment, and camera.duration: 6 / 8 / 10 seconds.generate_audio: true/false (default true).The rate is $0.067 per second.
Streamline scene design with high-fidelity, auto-interpolated video
Transform scripts or voices into dynamic, brand-tailored avatar videos fast.
Cinematic motion model for fluid scene creation and adaptive visual editing.
Seedance 2.5: Animate a still image into cinematic AI video
Turn images and text into motion-accurate HD videos fast.
Convert static visuals into seamless motion clips with audio control.
It animates a single reference image into a coherent video clip guided by your prompt. Use it to animate images with LTX 2 when you need cinematic motion, strong subject preservation, and optional synchronized audio.
This Pro I2V workflow is image-conditioned. The image_url anchors composition and identity, while the prompt specifies motion and camera. This mechanism yields higher fidelity and stability than text-only inputs.
Pro prioritizes fidelity over raw speed. Output is 1080p.
Resolution 1080p, durations of 6/8/10 seconds, aspect ratio fixed at 16:9.
First generate a solid base clip here, then refine specific regions, motion intensity, or timing in Retake: LTX 2 Retake Video. This avoids starting over and speeds up creative iteration.
Yes. Set generate_audio=true (default) to receive a video with synchronized ambience and sound effects that match on-screen events.
The rate is $0.067 per second.
RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.
