LTX 2.3 Pro
7.1/10Editorial score
Higher-rated areas: Creative control, Audio & lip sync
Compare engines
LTX 2.3 Pro vs Luma Ray 2
See what each model can create. Compare quality, controls and price.
How to choose
Compare LTX 2.3 Pro and Luma Ray 2 by key specs, pricing, controls, and a scorecard across 10 criteria.
Luma Ray 2
6.7/10Editorial score
Higher-rated areas: Visual quality, Creative control
See the models in action
Independent examples, with different prompts and settings.
LTX 2.3 Pro
No published video for this model yet. Explore the available examples.
Explore examplesLuma Ray 2
No published video for this model yet. Explore the available examples.
Explore examplesCompare what matters
These are MaxVideoAI editorial assessments, not same-prompt test results. The dates below record score updates. MaxVideoAI sells access to the models it covers.
Score records last updated: LTX 2.3 Pro · Luma Ray 2
How we benchmark| Criterion · 0–10 | LTX 2.3 Pro | Luma Ray 2 |
|---|---|---|
| Follows your promptHow closely the result follows the prompt. | LTX 2.3 Pro7.6/10 | Luma Ray 27.1/10 |
| Visual qualityDetail, lighting and visible artifacts. | LTX 2.3 Pro7.4/10 | Luma Ray 28.3/10 |
| Natural movementHow naturally subjects and cameras move. | LTX 2.3 Pro7.5/10 | Luma Ray 26.9/10 |
| Human appearanceFaces, hands and body movement. | LTX 2.3 Pro7.3/10 | Luma Ray 27.4/10 |
| Audio & lip syncSpeech, sound and mouth movement in sync. | LTX 2.3 Pro7.8/10 | Luma Ray 2N/A |
| Creative controlHow much direction the model follows. | LTX 2.3 Pro7.9/10 | Luma Ray 27.8/10 |
6 of 10 criteria shown10 of 10 criteria shownCompare the 4 other criteriaHide additional criteria
| Criterion · 0–10 | LTX 2.3 Pro | Luma Ray 2 |
|---|---|---|
| Consistency over timeWhether details hold together over time. | LTX 2.3 Pro6.2/10 | Luma Ray 26.1/10 |
| Readable textHow readable generated text and interfaces are. | LTX 2.3 Pro6.5/10 | Luma Ray 25.8/10 |
| Multi-shot continuityContinuity between different shots. | LTX 2.3 Pro6.5/10 | Luma Ray 25.8/10 |
| Speed & reliabilityGeneration speed and reliability. | LTX 2.3 Pro7.8/10 | Luma Ray 26.8/10 |
Generate with
LTX 2.3 Pro
Generate with
Luma Ray 2
What will your video cost?
Compare prices by resolution. The final quote depends on your duration, mode and settings.
LTX 2.3 Pro
- Current price unavailable
The current quote could not be confirmed. Check your settings in the app before generating.
Luma Ray 2
- Current price unavailable
The current quote could not be confirmed. Check your settings in the app before generating.
Your settings. Your quote. Before you generate.
USD estimate · snapshot generated
Formats, inputs and controls
Check duration, resolution, supported inputs and controls. Open each model’s guide for its full feature set.
New clip · maximum
- LTX 2.3 Pro
- 10 s
- Luma Ray 2
- 9 s
Max resolution
- LTX 2.3 Pro
- 4K on T2V/I2V generate; Reframe 720p/1080p; other workflow-specific limits
- Luma Ray 2
- 1080p
Audio output
- LTX 2.3 Pro
- Supported
- Luma Ray 2
- Not supported
Compare all 17 specificationsClose the full specification tableCheck references, audio, camera controls and output formats.
| Key spec | LTX 2.3 Pro | Luma Ray 2 |
|---|---|---|
| Pricing (MaxVideoAI) | Current price unavailable | Current price unavailable |
| Text-to-Video | Supported | Supported |
| Image-to-Video | Supported | Supported |
| Video-to-Video | Supported (Extend / Retake / Reframe) | Supported (modify / reframe workflows) |
| First/Last frame | Supported (start + end image in I2V) | Not supported |
| Reference image / style reference | Supported (single start image; no separate style-reference stack) | Supported (single start image) |
| Reference video | Supported (Extend / Retake / Reframe) | Supported (source clip for modify / reframe) |
| Max resolution | 4K on T2V/I2V generate; Reframe 720p/1080p; other workflow-specific limits | 1080p |
| New clip · maximum | 10 sText to video, Image to video: 6 / 8 / 10 s | 9 sText to video, Image to video: 5 / 9 s |
| Aspect ratios | 16:9 generate / 9:16 generate | 16:9 / 9:16 / 1:1 / 4:3 / 3:4 / 21:9 / 9:21 |
| FPS options | 24 fps generate / 25 fps generate / 48 fps generate / 50 fps generate | 24 |
| Output format | MP4 | Data pending |
| Audio output | Supported | Not supported |
| Native audio generation | Supported | Not supported |
| Lip sync | Supported | Data pending |
| Camera / motion controls | Prompt-based only | Not supported |
| Watermark | No (MaxVideoAI) | No (MaxVideoAI) |
Duration choices depend on mode and resolution; check the creation settings.
Capabilities available in MaxVideoAI. Provider sources below; their verification dates are not recorded here.
LTX 2.3 Pro: fal.ai (1) · fal.ai (2) · fal.ai (3) · fal.ai (4) · fal.ai (5) · fal.ai (6)
Luma Ray 2:
11 questions to help you choose
FAQ
Quick answers about LTX 2.3 Pro vs Luma Ray 2 on MaxVideoAI (pricing, modes, specs, and why results differ).
Open a question to read the answer. One answer at a time.
Which is better: LTX 2.3 Pro or Luma Ray 2?
It depends on your workflow. Use the scorecard and specs to compare control, references, audio, pricing, and generation limits, then open each engine profile for full details.
Which is cheaper on MaxVideoAI?
- LTX 2.3 Pro — Current quote unavailable; check settings before generating..
- Luma Ray 2 — Current quote unavailable; check settings before generating..
- Compare equivalent settings. These are estimates in USD, not reserved quotes.
What are the biggest differences between LTX 2.3 Pro and Luma Ray 2?
- Reference image / style reference — LTX 2.3 Pro: supported (single start image; no separate style-reference stack); Luma Ray 2: supported (single start image).
- Max resolution: LTX 2.3 Pro is 4K on T2V/I2V generate; Reframe 720p/1080p; other workflow-specific limits vs Luma Ray 2 is 1080p.
Can I start from text, an image or a video?
- LTX 2.3 Pro — Text-to-Video: Supported; Image-to-Video: Supported; Video-to-Video: Supported (Extend / Retake / Reframe).
- Luma Ray 2 — Text-to-Video: Supported; Image-to-Video: Supported; Video-to-Video: Supported (modify / reframe workflows).
Can I use start and end frames or references?
- LTX 2.3 Pro — First/Last frame: Supported (start + end image in I2V); Reference image / style reference: Supported (single start image; no separate style-reference stack); Reference video: Supported (Extend / Retake / Reframe).
- Luma Ray 2 — First/Last frame: Not supported; Reference image / style reference: Supported (single start image); Reference video: Supported (source clip for modify / reframe).
What resolutions, durations and aspect ratios are available?
- LTX 2.3 Pro — Max resolution: 4K on T2V/I2V generate; Reframe 720p/1080p; other workflow-specific limits; New clip · maximum: 10 s. Text to video, Image to video: 6 / 8 / 10 s. Aspect ratios: 16:9 generate / 9:16 generate. Duration choices depend on mode and resolution; check the creation settings.
- Luma Ray 2 — Max resolution: 1080p; New clip · maximum: 9 s. Text to video, Image to video: 5 / 9 s. Extension · added duration: Not supported. Aspect ratios: 16:9 / 9:16 / 1:1 / 4:3 / 3:4 / 21:9 / 9:21. Duration choices depend on mode and resolution; check the creation settings.
Can the models generate audio and lip sync?
- LTX 2.3 Pro — Audio output: Supported; Native audio generation: Supported; Lip sync: Supported.
- Luma Ray 2 — Audio output: Not supported; Native audio generation: Not supported; Lip sync: still being validated.
Does MaxVideoAI add a watermark?
- LTX 2.3 Pro — Watermark: No (MaxVideoAI).
- Luma Ray 2 — Watermark: No (MaxVideoAI).
Why can results differ between these models?
Models interpret instructions, visual references, and generation constraints differently.
Where can I find full specs, controls, and more prompt examples?
Open the full engine profiles for complete specs, controls, and more prompts: /models/ltx-2-3-pro and /models/luma-ray-2. You can also browse more outputs in the engine galleries.
What are LTX 2.3 Pro and Luma Ray 2?
LTX 2.3 Pro and Luma Ray 2 are AI video generation engines documented on MaxVideoAI. This page compares key specs, pricing, controls, and performance data shown above.