COMMUNITY PAGE
Run Topaz on Floyo
Home / Model / Topaz Video AI on Floyo
AI VIDEO UPSCALING
Run Topaz Video AI on Floyo
Professional video upscaling, denoising, sharpening, and frame interpolation. Upscale to 4K/8K with 19+ specialized AI models. No Topaz desktop license needed.
Run Topaz Video AI through ComfyUI in your browser. No installs, no local GPU, no desktop app required.
|
Upscale Up to 4K / 8K / 16K |
AI Models 19+ (Proteus, Apollo, Iris, etc.) |
|
Output H265/H264 MP4 with audio |
Speed ~25 sec per clip on Floyo |
| Try Topaz Video AI Now → | Browse All Models |
No installation. Runs in browser. Updated April 2026.
What you get?
Topaz Video AI is a professional video enhancement suite with 19+ specialized AI models for upscaling, denoising, sharpening, frame interpolation, stabilization, and motion deblur. Upscale from SD to 4K, 8K, or 16K. Recover detail from compressed, noisy, or low-resolution footage. Frame interpolation creates smooth slow motion or converts frame rates. On Floyo, it runs as a ComfyUI API node with no desktop license required. Upload a video, pick your model, and get enhanced H265 MP4 output with audio preserved.
TOPAZ VIDEO AI WORKFLOWS ON FLOYO
What is Topaz Video AI?
Topaz Video AI is a professional video enhancement suite from Topaz Labs. It uses 19+ specialized AI models to upscale, sharpen, denoise, interpolate frames, stabilize, and deblur video footage. Each model is trained for a specific task: Proteus for general enhancement, Iris for face recovery, Nyx for low-light denoising, Apollo for frame interpolation, and more. It can upscale video from SD to 4K, 8K, or even 16K.
The desktop software normally costs $299/year. On Floyo, you access Topaz Video AI through a ComfyUI API node with no desktop license needed. Upload your video, pick the enhancement model and quality level, and Topaz processes it on cloud infrastructure. Output comes back as H265 MP4 with your original audio preserved.
Topaz Video AI works best on real-world footage. Interviews, product shots, drone footage, archival material, and anything that looks soft, compressed, or low-resolution gets significant improvement. Old 1080p footage that needs to look good at 4K, or 720p clips from older cameras, see the biggest gains. For AI-generated video, results vary because the models are trained on real-world content.
On Floyo, the workflow handles encoding and frame assembly for you. The Topaz API node processes each frame, and the VHS_VideoCombine node reassembles the enhanced frames into a final video file. Generation takes about 25 seconds per clip on Floyo's infrastructure.
What are Topaz Video AI's technical specifications?
Topaz Video AI includes 19+ AI models organized by task: enhancement (Proteus, Iris, Nyx, Rhea, Starlight), frame interpolation (Apollo, Chronos, Aion), stabilization, and motion deblur. Upscales up to 4x (from any resolution to 4K/8K/16K). Output supports H265/H264 in MP4/MOV/ProRes. Audio is preserved through the pipeline. On Floyo, it runs as the TopazVideoEnhance API node.
| Spec | Details |
|---|---|
| Developer | Topaz Labs |
| AI Models | 19+ specialized models (Proteus, Iris, Nyx, Rhea, Starlight, Apollo, Chronos, Aion, Artemis, and more) |
| Default Model | Proteus v4 (prob-4) for general sharpening and detail recovery |
| Upscale | Up to 4x (SD to 4K, 8K, or 16K) |
| Frame Interpolation | Apollo 8 (default), Chronos, Aion for smooth slow motion and frame rate conversion |
| Tasks | Upscaling, denoising, sharpening, frame interpolation, stabilization, motion deblur, deinterlacing |
| Output Codecs | H265 (default), H264, ProRes |
| Output Format | MP4 (default), MOV |
| Audio | Preserved (AAC default) |
| Quality Levels | Low / Mid (default) / High |
| Speed on Floyo | ~25 seconds per clip |
| ComfyUI Node | TopazVideoEnhance_floyo (Floyo API node) |
| Desktop License | Not required on Floyo (runs as API) |
What can you create with Topaz Video AI?
Topaz Video AI covers video upscaling to 4K/8K/16K, noise reduction, compression artifact removal, face recovery, frame interpolation for slow motion, frame rate conversion, video stabilization, motion deblur, and deinterlacing. It is the final step in production pipelines where generated or shot footage needs polish before delivery.
| Capability | What It Does | Use Case |
|---|---|---|
| 4K/8K Upscaling | Upscale from SD, 720p, or 1080p to 4K, 8K, or 16K. Rebuilds lost pixels, textures, and clarity using AI models trained on diverse footage. | Archival restoration, broadcast delivery, large-format displays |
| Denoising | Remove grain, sensor noise, and visual disturbances. Nyx for low-light, Proteus for general, Iris for faces. | Low-light footage, high-ISO material, old recordings |
| Artifact Removal | Clean up compression artifacts, banding, blocking, and JPEG degradation from over-compressed source files. | Web-sourced footage, social media re-exports, old DVD rips |
| Frame Interpolation | Create smooth slow motion or convert frame rates (24 to 60fps, etc.). Apollo 8 handles non-linear motion. Aion handles 4K+ with fewer artifacts. | Cinematic slow motion, sport replays, frame rate conversion |
| Face Recovery | Recover facial detail from low-res or blurry footage. Iris model specializes in skin texture and facial feature restoration. | Interview footage, talking head content, surveillance enhancement |
| Pipeline Finishing | Chain as the final step after AI video generation. Generate with Wan 2.7 or Seedance, then upscale and sharpen with Topaz in the same ComfyUI workflow. | Production pipelines, AI video polishing, delivery prep |
What are Topaz Video AI's key features?
Topaz Video AI's feature set is modular: choose the right model for your footage type, and the output quality follows. The models are trained on specific content types (faces, low-light, animation, general footage) rather than being one-size-fits-all. This specialization is why Topaz produces cleaner results than generic upscalers.
Proteus v4 (General Enhancement)
The default model for most footage. Handles sharpening, noise reduction, and detail recovery in one pass. Allows manual fine-tuning of enhancement parameters: adjust the balance between detail recovery and noise reduction based on your source. Start with Proteus for any footage type, then switch to specialized models if needed.
Apollo 8 (Frame Interpolation)
Creates smooth slow motion or converts frame rates. Optimized for non-linear motion like handheld camera shake. If your source has choppy movement or low frame rates, Apollo fills in the gaps. For 4K+ footage and large motions, Aion (the 2026 flagship interpolator) produces fewer tiling artifacts.
Iris and Nyx (Specialized Recovery)
Iris focuses on face recovery: skin textures, facial features, and detail in close-up shots. Nyx specializes in low-light denoising: footage shot in dark environments or at high ISO settings. Both preserve natural-looking results without the over-sharpened artificial look that generic enhancement tools produce.
Starlight (Diffusion Upscaler)
The newest model in the Topaz family. A diffusion-based upscaler that transforms low-resolution footage into clean HD/4K without manual controls. It generates realistic fine details rather than just sharpening existing pixels. This is the first diffusion model in Topaz that can run locally.
Audio Preservation
Unlike many upscaling tools that strip audio, Topaz on Floyo preserves your original audio track through the pipeline. AAC codec by default. The output is a complete MP4 file ready for delivery, not a silent video that needs audio re-attached.
No Desktop License Required on Floyo
Topaz Video AI desktop normally costs $299/year. On Floyo, you access it through the TopazVideoEnhance API node. No desktop app download, no license key, no local GPU requirement. The API handles processing on cloud infrastructure. You pay through your Floyo API Wallet instead of a separate Topaz subscription.
How does Topaz Video AI work?
Topaz Video AI uses machine learning models trained on diverse video datasets. Each model is specialized: Proteus is trained for general enhancement and detail recovery, Iris for facial features, Nyx for low-light conditions, Apollo for temporal interpolation. The models analyze each frame, predict what the original high-resolution detail should look like, and reconstruct it.
For upscaling, the model takes a low-resolution frame and generates the missing pixels based on learned patterns. This is different from traditional upscaling (bicubic, Lanczos) which just smooths between existing pixels. The AI predicts textures, edges, and fine detail that were lost during compression or downscaling.
For frame interpolation, models like Apollo analyze motion between existing frames and generate new in-between frames. This creates smooth slow motion or converts a 24fps source to 60fps. The motion prediction handles non-linear camera movements and complex object interactions better than optical flow methods.
On Floyo, the workflow is: LoadVideo node imports your file, TopazVideoEnhance_floyo sends each frame to the Topaz API for processing, VideoToFrames handles frame extraction, and VHS_VideoCombine reassembles the enhanced frames with your original audio into a final H265 MP4. The entire pipeline runs in about 25 seconds per clip on Floyo's infrastructure.
Fair warning: Topaz models are trained on real-world footage. AI-generated video with unusual artifacts or flat textures may not always improve. Stylized content (anime, illustration) can produce mixed results. Always test a short clip before committing to a long render. Processing time scales with resolution and duration: longer, higher-resolution footage takes proportionally longer.
Frequently Asked Questions
Common questions about running Topaz Video AI on Floyo.
You can start with Floyo's free pricing plan. Floyo gives $0.25 in free API credits on signup. To continue using the service beyond the free tier, upgrade your Floyo pricing plan. Topaz Video AI runs as an API node, so processing costs come from your API Wallet (separate from your plan's GPU time). No $299/year Topaz desktop license required.
Open Floyo in your browser, find the "Topaz Video Upscaler for Sharper Results" workflow (search "Topaz" in the template library), and click Run. Upload your video, pick the enhancement model, set quality level, and generate. Floyo handles the ComfyUI environment and API connection. No desktop app, no Python setup, no local GPU.
Topaz Labs, a company specializing in AI-powered photo and video enhancement tools. The desktop software has been a professional standard for video upscaling and restoration since its initial release. The 2026 version (Video AI 7) includes 19+ models including the new Starlight diffusion upscaler and Aion frame interpolator.
Start with Proteus v4 (prob-4). It handles sharpening, noise reduction, and detail recovery for most footage. Switch to Iris for face-heavy content, Nyx for low-light footage, or Starlight for very low-resolution sources that need generated detail. For frame interpolation, Apollo 8 is the default. Test a short clip before processing long videos.
It can, but results vary. Topaz models are trained on real-world footage. AI-generated content with unusual artifacts or flat textures may not improve, or may introduce unwanted changes. For AI video polishing, run a short test clip first. Products shots and photorealistic AI video tend to work better than highly stylized content.
Yes. That's the main advantage of running Topaz in ComfyUI on Floyo. Generate video with Wan 2.7, Seedance 2.0, or Kling Omni, then pass the output to Topaz for upscaling and sharpening in the same workflow. The Topaz node sits at the end of your pipeline as the finishing step before final export.
Yes. The workflow preserves your original audio track. Output is a complete H265 MP4 file with audio. AAC codec by default. You get a ready-to-deliver file, not a silent video that needs audio re-attached.
No. On Floyo, Topaz Video AI runs as a cloud API node (TopazVideoEnhance_floyo). Processing happens on Topaz's servers, accessed through your Floyo API Wallet. You don't need the $299/year desktop app, a Topaz account, or a local GPU. Everything runs in your browser through the ComfyUI workflow.
Try Topaz Video AI on Floyo
Professional video upscaling, denoising, and frame interpolation with 19+ AI models. No desktop license needed. Run it in your browser.
| Try Topaz Video AI Now → | Browse All Models |
Related Reading
VFX and Post Production on Floyo
Film and Animation Workflows on Floyo
Last updated: April 2026. Specs from Topaz Labs official documentation, Topaz Video AI presets guide, third-party reviews, and Floyo workflow page.
%20(2)_1774208003255.webp?width=400&height=300&quality=80&resize=cover)
API
Floyo API
Topaz
Video2Video
Video Upscale
Upload a video, pick your enhancement model and quality level, and Topaz Video AI sharpens, denoises, and upscales it. Audio is preserved. Output is H265 MP4.
Topaz Video Upscaler for Sharper Results
Upload a video, pick your enhancement model and quality level, and Topaz Video AI sharpens, denoises, and upscales it. Audio is preserved. Output is H265 MP4.