Open Models for Flexible Workflows

LTX-2 is our flagship open-source video model with native audio sync. This page lists models tagged as Open for workflows that benefit from flexible generation.
The tag is a practical signal of flexibility, not a permission slip—content rules still apply.

Open AI Video Models

Generate videos from text or images. Start with LTX-2 for native audio sync, then explore other Open video models for different styles and trade-offs.

Wan 2.6

Wan 2.6

NEW
Next-gen Wan AI model with enhanced video quality and features.
🌶🌶🌶🌶🌶Unrestricted720P/1080P: 30/45 Credits/sReference to Video: 60/90 Credits/sImage to VideoText to VideoAudio Synchronization
Wan 2.5

Wan 2.5

Latest Wan AI model with audio synchronization.
🌶🌶🌶🌶🌶Unrestricted1080P: 120 credits/3s720P: 80 credits/3s480P: 40 credits/3sImage to VideoText to VideoAudio Synchronization
Wan 2.2

Wan 2.2

Professional AI video generation with 480p/720p resolution support. Text-to-video and image-to-video.
🌶🌶🌶🌶🌶Unrestricted720p: 90 credits/5s, 144 credits/8s480p: 45 credits/5s, 72 credits/8sText to VideoImage to Video
Wan 2.2 Spicy

Wan 2.2 Spicy

Affordable Grok Imagine alternative - High dynamics & smooth motion at 80% lower cost. No subscription needed.
🌶🌶🌶🌶🌶Unrestricted720P: 75 credits/5s, 120 credits/8s480P: 40 credits/5s, 64 credits/8sImage to Video
Wan 2.2 Animate

Wan 2.2 Animate

Face/Body Swap and also Copy Motion.
🌶🌶🌶🌶🌶Unrestricted720P: 120 credits/5s480P: 60 credits/5sImage to VideoText to Video
Kling 2.6 Pro

Kling 2.6 Pro

NEW
Premium Kling 2.6 model with optional audio and motion control.
🌶🌶🌶Unrestricted5s: 105 credits (no audio)10s: 210 credits (no audio)5s: 210 credits (with audio)10s: 420 credits (with audio)Motion Control: 32 credits/sText to Video / Image to Video / Motion Control
Hailuo 2.3

Hailuo 2.3

Transform images into dynamic videos with smooth motion and realistic animation.
🌶🌶🌶UnrestrictedStandard(768P, 6/10s): 90/180 creditsPro(1080P, 5s): 150 creditsImage to VideoText to Video
LTX-2 19B

LTX-2 19B

NEW
DiT-based audio-video foundation model with synchronized audio and high fidelity. 5-20 seconds generation.
🌶🌶🌶Unrestricted480p: 4 credits/s720p: 6 credits/s1080p: 8 credits/s5-20 secondsText to VideoImage to Video16:9 / 9:16
Seedance V1 Pro

Seedance V1 Pro

ByteDance's advanced video generation model with multi-resolution support and First-Last Frame mode.
🌶🌶🌶🌶Unrestricted480p/720p/1080p9/18/36 credits/s2-12 secondsFirst-Last FrameText to VideoImage to Video
Seedance V1 Pro Fast

Seedance V1 Pro Fast

ByteDance's fast video generation model with 50% lower cost. I2V mode requires prompt, no Last Frame support.
🌶🌶🌶🌶Unrestricted480p/720p/1080p4/8/16 credits/s2-12 secondsText to VideoImage to Video
Seedance V1.5 Pro

Seedance V1.5 Pro

NEW
ByteDance's latest video model with AI audio generation, video extend, and First-Last Frame mode.
🌶🌶🌶🌶Unrestricted480p/720p4-8 / 8-16 credits/s (audio doubles)4-12 secondsT2V / I2V / Video ExtendFirst-Last FrameAudio Generation
Seedance V1.5 Pro Fast

Seedance V1.5 Pro Fast

NEW
Fast variant with higher resolution (720p/1080p) and optimized generation speed.
🌶🌶🌶🌶Unrestricted720p/1080p6-9 / 12-18 credits/s (audio doubles)4-12 secondsT2V / I2V / Video ExtendFirst-Last FrameAudio Generation
Video Upscaler

Video Upscaler

Upscale videos to 4K with three quality options: Upscaler (affordable), Upscaler Pro (best quality), and FlashVSR (balanced speed).
🌶🌶🌶🌶🌶UnrestrictedUpscaler: 1-4 credits/sUpscaler Pro: 4-10 credits/sFlashVSR: 2-8 credits/sUp to 4K resolutionMinimum 5s billing

Open AI Image Models

Create reference images, concepts, and keyframes—then bring them to life with LTX-2 image-to-video.

Wan 2.5 Image

Wan 2.5 Image

Professional image generation and editing with Wan 2.5 AI model.
🌶🌶🌶🌶🌶Unrestricted10 credits per imageText to ImageImage to Image1K
Qwen Image

Qwen Image

Advanced 20B MMDiT model for high-quality image generation and multi-image editing with native ControlNet support.
🌶🌶🌶🌶🌶Unrestricted8 credits per imageText to ImageMulti-Image Edit (1-3 images)1K
Z-Image

Z-Image

NEW
Alibaba's 6B parameter image model enable seconds-level image generation.
🌶🌶🌶🌶🌶Unrestricted3 credits per imageTurbo (T2I)Image editing on the way1K
Seedream 4.5

Seedream 4.5

NEW
ByteDance's latest 4K text-to-image model optimized for typography—crisper text rendering and stronger prompt adherence
🌶🌶🌶🌶Unrestricted12 credits per imageText to ImageImage to Image4KText Rendering
Seedream-V4

Seedream-V4

ByteDance's state-of-art image generation model with up to 4K resolution.
🌶🌶🌶🌶Unrestricted8 credits per imageText to ImageImage to Image4K
GPT Image 1.5

GPT Image 1.5

NEW
GPT Image 1.5 is a new multimodal image generator with natural language understanding.
🌶🌶🌶UnrestrictedLow: 3-4 creditsMedium: 10-15 creditsHigh: 40-60 creditsT2I/Edit(up to 10 images)1024×1024 / 1024×1536 / 1536×1024
Flux-2

Flux-2

NEW
FLUX.2 from Black Forest Labs - Fast image generation and editing with multiple quality tiers.
🌶🌶🌶UnrestrictedKlein 4B: 3/4 creditsKlein 9B: 3/5 creditsDev: 4/8 creditsPro: 9/18 creditsFlex: 18/36 creditsMax: 9/18 creditsText to ImageMulti-Image Edit (1-3 images)256×256 - 1536×1536
Flux1-Kontext

Flux1-Kontext

Advanced image generation with exceptional prompt understanding and editing capabilities. Trained by Black Forest Labs.
🌶🌶🌶UnrestrictedPro: 8 creditsMax: 15 creditsText to ImageImage EditingMultiple aspect ratios

Pricing

🎊 Begin creating with our AI image and video models! Select a credits package to get started. 🎊 Prices shown exclude VAT. Applicable taxes vary by location and will be calculated at checkout.

Flexible credit purchases - acquire credits as needed, with no expiration date.

Basic

$10one-time

Ideal for exploring AI video creation.

Includes

  • 800 credits
  • Up to 20 video generations
  • Up to 100 image generations
  • Credits never expire
  • HD quality download
  • No watermark output
  • Commercial use allowed

Excellent for experimentation and compact projects!

Popular

Popular
$28$25one-time

Top pick among video creators.

Includes

12% More Credits
  • 2200 credits
  • Up to 55 video generations
  • Up to 275 image generations
  • Credits never expire
  • HD quality download
  • No watermark output
  • Commercial use allowed
  • Priority support

Optimal value for frequent creators!

Pro

$62$50one-time

Designed for professionals and commercial use.

Includes

20% More Credits
  • 5000 credits
  • Up to 125 video generations
  • Up to 625 image generations
  • Credits never expire
  • HD quality download
  • No watermark output
  • Commercial use allowed
  • Priority support
  • Bulk generation capabilities

Tailored for production-grade workflows!

How to Use LTX-2 and Open Models

LTX-2 is an open-source video model designed for production-ready generation with native audio sync. This page highlights models tagged as Open for flexible creative workflows. Use responsibly and follow our Terms and applicable law.

FAQ

Open Models FAQ

Common questions about LTX-2 and models tagged as Open on ltx-2.pro.

1

What does the 'Open' (Unrestricted) tag mean here?

It highlights models that tend to be more flexible in practice. You still must follow our Terms of Service and applicable laws, and some prompts may be rejected based on safety rules.

2

Is LTX-2 open-source and can I run it locally?

Yes. LTX-2 is open-source, and local runs can use quantized versions on high-end consumer GPUs. If you prefer zero setup, use the web generator.

3

How can I control camera motion and style?

Use explicit camera language in the prompt. In local workflows, LoRA-based camera controls (dolly, pan, truck) can make motion more predictable and repeatable.

4

How does pricing work across different models?

All models on ltx-2.pro use a unified credit system. Purchase credits once and use them across LTX-2 and other supported models. No separate accounts or subscriptions needed.View pricing.

5

Can I use outputs for commercial projects?

Yes, subject to our Terms of Service. Make sure your use complies with local advertising rules, platform policies, and any rights or licensing requirements.

6

Why do some models refuse certain prompts?

Different model providers apply different safety filters and capabilities. If a request fails, try a different model or rewrite the prompt to be clearer and within policy.

Try LTX-2 on ltx-2.pro

Generate video with native audio sync, longer coherent clips, and directable motion—then explore other Open models for different creative needs.