Dialogue and lip-sync
“Close-up of a barista saying, "Your order is ready," then smiling as the espresso machine hisses behind her.”
Check speech clarity, mouth timing, facial stability, and whether the background sound fits.
LTX 2.3: Features & Alternatives
LTX 2.3 has been replaced by LTX-2.5 on Magic Hour. Use LTX-2.5 for new projects; the information below describes the earlier model.
Can't park there
Deadpool
Funny friend
Outreach tip
Singer
Spongebob
Will Smith
How to evaluate LTX 2.3
A practical decision guide based on Magic Hour's current product support and the questions creators ask when comparing models.
Best fit on Magic Hour
Fast video iteration when native audio, spoken dialogue, and flexible clip lengths matter as much as the picture.
Consider another model when
Compare a premium video model when demanding fast motion or maximum cinematic polish matters more than iteration speed and cost.
Three useful first-run tests
Use the prompts below to test the failure modes that matter before committing to a larger batch.
Prompt-detail sensitivity
“Run the same scene once as a short request and once with subject, action, camera, lighting, dialogue, and ambient sound described.”
Compare instruction coverage and whether added detail improves or confuses the scene.
Motion under occlusion
“A skateboarder crosses behind a parked van, reappears, lands a turn, and rolls toward camera as traffic passes.”
Watch identity, limbs, direction of travel, and audio continuity through the obstruction.
These prompts are evaluation recipes, not controlled benchmark results. Run the same prompt and settings across models before drawing a conclusion.
Trusted by teams at
How to Make Videos with LTX-2.5
LTX 2.3 has been replaced. Follow these steps with LTX-2.5.
- 1
Type a Prompt
Type a word or set of words in the text field. You can combine several words with or without commas. - 2
Customize Video
Optionally upload an image to use as the first frame, change the aspect ratio or duration, or select a template. - 3
Generate Your Video
Click "Generate Video" to start. Once it's ready, download your high-quality video in your preferred format.
What LTX 2.3 Was Known For
Where LTX 2.3 really shines: fast iteration, synchronized audio-video generation, expressive faces, and practical control for teams that want to prototype, test, and ship without overpaying for every draft.
Fast iteration
Generate drafts quickly when you need to test prompts, pacing, camera ideas, or multiple creative directions in one session.
Built-in audio workflows
LTX 2.3 is one of the clearest options when you want synchronized audio and video in the same generation pipeline.
Expressive faces and lip-sync
Useful for characters, talking shots, and reference-led clips where facial expression matters.
No Editing Skills Required
Turn your ideas into polished outputs quickly — no complex editing workflow needed. Just use a clear prompt, a reference if helpful, and iterate.
Thousands of Templates
Make your photos even better and faster with our high-quality templates.
Explore TemplatesHow LTX 2.3 Compares to Other Video Models
LTX 2.3 THIS MODEL | Veo 3.1 | Kling 3.0 | MiniMax H3 | Seedance 2.5 | Kling 2.6 | |
|---|---|---|---|---|---|---|
Best for | Historical model; replaced in Magic Hour | Premium realism, polish, dialogue, and strong prompt adherence | Multi-shot storytelling, character references, structured cinematic control | Reference-driven clips with native audio, character consistency, longer short-form | Cinematic continuity, structured references, narrative short-form | Action, motion blur, and controlled camera movement |
Visual fidelity | Good | Premium realism | Strong | Strong | Strong | Good |
Real-human image to video | Good with the right source image | Better suited | Better suited | Better suited | Good | Supported |
Viral clips | Medium | Good | Strong | Strong | Strong | Motion-driven clips |
Multi-scene storytelling | Medium | Strong | Excellent | Medium to strong | Strong | Short clips up to 10s |
Surreal / imaginative concepts | Medium | Strong | Good | Good | Strong | Prompt-led generation |
Native audio | Historical feature | Yes | Yes | Yes | Yes | No |
Template fit | Good | Good | Good | Good | Good | Good |
What LTX 2.3 Was Known For
Where LTX 2.3 really shines: fast iteration, synchronized audio-video generation, expressive faces, and practical control for teams that want to prototype, test, and ship without overpaying for every draft.
Fast ideation
One of the best models here for running many prompt variations before you commit to a final concept.
Native synchronized audio
Audio comes with the generation, which reduces handoff friction when you want more finished-looking drafts.
Lip-sync friendly clips
Helpful for talking shots, expressive characters, and performances where the mouth and face matter.
Flexible duration options
Supports short social clips and longer tests up to 30 seconds, which is useful for storyboard-style exploration.
Open, controllable workflows
Strong fit for teams that value a more production-oriented pipeline and want a model with open access roots.
Cost-efficient experimentation
Often a better choice than premium video models when your main constraint is iteration speed.
LTX 2.3 — Historical Model Card
Key specs, capabilities, and limitations.
Cost
Historical: 24–72 credits/sec depending on resolution
Resolution
480p, 720p, 1080p
Aspect ratios
9:16, 16:9, 1:1
Max duration
Up to 30s
Audio
Yes, native
Plans
No current LTX 2.3 plan
Limitation
Not the strongest option for absolute top-end cinematic fidelity
Frequently asked questions
No. LTX 2.3 has been replaced by LTX-2.5 in Magic Hour. The earlier model offered synchronized audio-video generation, expressive faces, and fast iteration.
Historically, LTX 2.3 was used for quick drafts, audio-video experiments, lip-sync friendly shots, and expressive characters. Use LTX-2.5 for new generations.
Historically, LTX 2.3 supported 480p, 720p, and 1080p output, 9:16, 16:9, and 1:1 aspect ratios, built-in audio, and durations up to 30 seconds.
Historically, yes. LTX 2.3 included synchronized audio natively, which is one of its clearest advantages versus many other video models.
Historically, yes. LTX 2.3 previously supported both text-to-video and image-to-video style workflows, which makes it useful for animating still images as well as generating from scratch.
LTX 2.3 is no longer offered. Its replacement, LTX-2.5, costs 24 credits/sec at 480p, 48 at 720p, and 72 at 1080p with native audio included.
LTX 2.3 is no longer a current model option. Its historical strengths were speed, control, and integrated audio; compare available models on your own brief.
Historically, it could animate a person from a clear source image. For new work, evaluate LTX-2.5 or another available image-to-video model.
LTX 2.3 previously supported 9:16, 16:9, and 1:1 inside Magic Hour.
Use LTX-2.5 for new work. Magic Hour lets you compare it with Kling, Veo, Seedance, and MiniMax in one place.
Choose LTX-2.5 for fast audio-video iteration, uploaded audio, and clips up to 60 seconds. Its 480p setting is available on all plans; 720p and 1080p require paid access.
Paid subscriptions and credit-pack purchases include commercial-use rights. Users without either purchase are limited to personal, non-commercial use. Outputs created with LTX 2.3 on Magic Hour can be used commercially, subject to our Terms.
Related Models
Compare models and create from text, images, or clips with Magic Hour's AI video generator.