A complete Windows video editor with an AI agent beside your timeline. Organize footage, arrange tracks, refine titles and captions, balance music, and inspect the preview in the same conversation. Connect your own services for music, speech and lip-sync, or generate with supported local workflows and configured native WanGP presets. Choose the language model that drives the agent. Local model detection is separate from generation compatibility; API usage and model requirements depend on your setup. Your purchase includes private GitHub access to download the Windows installer.
View pricing“Keep both characters.
Move the scene to the river.”
This is LTX Desktop Enhanced, from Version Six. A complete Windows video editor, with an AI agent right beside your timeline. Bring your footage, shape your story, and keep your creative tools in one place. Let me show you how it works. Start with your own footage, images, and audio. Ask the agent to organize your assets into bins and arrange the edit across picture, music, and title tracks. Trim a shot. Split a clip. Adjust the pacing. Add a title, then change its wording, size, color, and position. Add captions, balance audio levels, and preview the result. You can keep refining the same project in the same conversation. For generated scenes, build a plan first. Reference multiple characters, define their looks, and create a new scene still before turning that scene into video. Review the stills and clips as you go, or choose your approval settings. LTX Desktop Enhanced also brings more generation choices into the editor. Connect your own services for music, speech, and lip-sync. Discover local image, video, and audio models, with clear setup and availability information. Use supported ComfyUI workflows, or run tested native presets through your existing WanGP installation. Your files and generated results come back into the same creative workflow. From a short story to a music video, a promotional ad, or a longer production, keep building scene by scene and shaping the timeline. Get LTX Desktop Enhanced from Version Six. One purchase gives you a lifetime software license and private GitHub access to download the complete Windows application. Your API usage and third-party services are separate. See the features, system requirements, and limited-time offer at version dash six dot com slash tools.
Watch LTX Desktop Enhanced in action.
A license buys read-only access to version6-tools/LTX-Desktop-Enhanced. That private repository is the download location for the Windows installer, updates, and optional source. You do not need to compile anything to use the editor.
No. It builds on the open-source LTX Desktop and is an independent Version 6 edition, not an official Lightricks product.
Yes. In version 1.2.11, type while the agent is working and press Enter to queue the next request. Choose Steer now or press Ctrl+Enter to interrupt the current run and prioritize your new direction. Stop pauses the queue; Resume continues it. Completed edits stay in the project. Queued prompts are held in the current session, and voice interruption is not included.
Yes. The agent can add, adjust, disable, and remove blur, cinematic, vintage, black-and-white, cool, warm, muted, and vivid looks. These effects work in preview and video export. It can also apply fades, paired cross-dissolves, wipes, and color correction. Effect masks, sharpen, glow, grain, vignette, and zoom transitions are not supported by this release.
The agent lives in a side window next to the LTX Desktop timeline. Ask it to split, trim, generate, assemble, narrate, or lip-sync a clip. It works on the same assets you can see, and it can use LTX or extra local models you already installed. Queue your next instruction while it works, or choose Steer now to change direction.
You are not limited to LTX 2.5. Settings → Models discovers installed image, video, music and speech models and explains their setup requirements. Use supported ComfyUI workflows or configured native WanGP presets. A detected checkpoint is not automatically ready to generate. Updates for licensed users come from the private GitHub Releases page in About.
Bind character references, generate scene stills, assemble onto the timeline, and review shot by shot. Add a sync.so API key in Settings to match a talking clip to dialogue or a song.
Lifetime software license. No recurring subscription. All sales are final. There are no refunds.
$299.99$199.99 USD No subscription
Save $100 on your lifetime license
The complete Windows LTX Desktop editor, with the video editor AI agent built in. After payment you receive read-only GitHub access to download the installer. No coding required. LLM keys, sync.so, extra local models through ComfyUI or Wan2GP, API credits and compatible hardware are separate.
A compatible Windows PC for the supplied installer. Local video generation requires a supported NVIDIA GPU, sufficient VRAM and model storage. Extra local models such as MiniMax H3 also need ComfyUI running on this PC. Wan2GP ckpts folders are scanned automatically when present. API generation and agent providers require your own accounts and may charge separately. Check the installation guide for your intended model before purchasing.
Yes. Settings → Models can scan ComfyUI, Wan2GP ckpts folders, and folders you add for MiniMax H3, Wan, Hunyuan, ACE-Step, MiniMax Music, Stable Audio, Chatterbox, FLUX, Qwen Image, SDXL, and other installed checkpoints. Detection does not guarantee compatibility. Generate through supported ComfyUI workflows or explicitly configured native WanGP presets; availability depends on matching weights, companion models, nodes and hardware. Built-in LTX and Z-Image Turbo remain available.
Yes. Describe footsteps, rain, traffic, impacts, whooshes or other sounds and ask the agent to place them at a specific time, or save them in your assets. ElevenLabs provides sound effects and optional looping ambience using your API key. Compatible local Stable Audio workflows can generate sound effects without that cloud service. MusicAPI and local ACE-Step create music; Chatterbox creates speech. Each model is used for its supported audio task. Local setup and hardware are required, provider charges are separate, and generated timing and sound quality need review.
No. The agent can send a talking clip and an audio asset to sync.so when you add a sync.so key in Settings. Review the mouth movement. Character identity and framing also need your eye.
The private GitHub repository version6-tools/LTX-Desktop-Enhanced. A paid license grants read-only access so you can download the installer from Releases.
You can plan and edit longer productions shot by shot. Feature-length generation has not been validated end to end; generation time, storage and manual review grow with the project.
A one-time license for the lifetime of this product, without a recurring software subscription. It excludes generation credits and third-party services. Refunds and payment disputes affect repository access as described in the product license.