New Features Real-time super-resolution and frame generation for various windows (players, web pages, games) Turbo Mode support for non-NVIDIA GPUs Next-generation WAFT VFI model replacing Tariff Optimizations Improved Discord status display Improved forum page display Other opti
New Features
[Pro Version] [Open Beta] Real-time super-resolution and frame generation for various windows (players, web pages, games)
[Pro Version] [Open Beta] Turbo Mode support for non-NVIDIA GPUs
[Pro Version] Next-generation WAFT VFI model replacing Tariff
Optimizations
Improved Discord status display
Improved forum page display
Other optimizations
Fixes
Fixed jittering issues with Tariff VFI model in certain scenarios
SVFI Open Beta Release Compared to the stable release, the current open beta includes: Real-time super-resolution and frame generation for various windows (players, web pages, games) Turbo Mode support for non-NVIDIA GPUs, based on DirectML backend, compatible with .onnx super-re
SVFI Open Beta Release
Compared to the stable release, the current open beta includes:
[Pro Version] Real-time super-resolution and frame generation for various windows (players, web pages, games)
[Pro Version] Turbo Mode support for non-NVIDIA GPUs, based on DirectML backend, compatible with .onnx super-resolution models under the TensorRT category, as well as RIFE models
Instructions for real-time super-resolution and frame generation:
To apply VFI or super-resolution to any window, first change "Auto Scale" in the Capture section from "Off" to "Windowed Scaling" or "Fullscreen Scaling"
Select the target window, e.g., a browser or media player. Scaling of Browsers currently do not support DRM-protected content. Scaling full-window games (such as Zenless Zone Zero) is not supported either.
Set capture method to default "Graphics Capture". If stuttering occurs, try switching to "GDI".
Set scaling mode to default "Lanczos". For AI super-resolution, select "ONNX" then choose a lightweight 2x super-resolution model from the dropdown (e.g., .onnx models with "SuperUltra" in the name)
For display strategy, if you want the window size to remain unchanged (e.g., frame generation only), select "Don't resize". Otherwise, select "Resize by Ratio" to enlarge the window proportionally.
For scaling strategy, when using methods with significant visual changes (e.g., AI super-resolution), it is recommended to keep the default "Scale and Downscale"
Check "Enable Frame Generation" for smoother window performance
When using frame generation for the first time, be sure to select the ABME algorithm. It is recommended to match the capture framerate with the playback/game framerate — e.g., 24 FPS for anime, minimum 60 FPS for games with locked framerates. By configuring different combinations of frame generation multipliers and capture framerates, you can achieve various target framerates. Ideally, set it to an integer multiple of your monitor's refresh rate for vsync.
For better frame generation quality, consider selecting RIFE. For AMD GPUs or NVIDIA GPUs where you want to avoid compilation/preprocessing time, set the inference backend to DirectML. For NVIDIA GPUs, you can select TensorRT, but the model will undergo pre-compilation first, requiring a wait of approximately 10–30 minutes (the progress bar at the top will show "Preprocessing"). Once completed, you will see the frame generation overlay window. Note that RIFE models will recompile once at resolutions above 2K. Generally, rife 4.6 and rife 4.8 are recommended. Devices below the RTX 40 series should choose rife 4.6, otherwise performance may suffer.
upload log/realtime_pyhook.log if the window still can not be scaled at software's internal forum to help us diagnose the problem.
Interested users, please navigate to SVFI library page → Properties → Betas tab, and select "beta" from the dropdown to join the open beta.
New Features Added Waft frame interpolation model Smooth Diff Fluency Optimization mode, specifically designed for duplicate frame removal at 1x frame rate output; supports Turbo Mode NVIDIA Fast Encoding now supports high-precision workflows Player now supports previewing more t
New Features
[Professional Edition] Added Waft frame interpolation model
Smooth Diff Fluency Optimization mode, specifically designed for duplicate frame removal at 1x frame rate output; supports [Professional Edition] Turbo Mode
NVIDIA Fast Encoding now supports high-precision workflows
Player now supports previewing more types of inputs
Embedded discussion community now includes "Mark as Read" option
New Features Turbo Mode now supports scene detection extrapolation Turbo Mode super-resolution now supports tiling inference Tariff frame interpolation models compatible with non-NVIDIA GPUs New achievements and leaderboard display interface Optimizations Default decoding mode (w
New Features
[Professional Edition] Turbo Mode now supports scene detection extrapolation
[Professional Edition] Turbo Mode super-resolution now supports tiling inference
[Professional Edition] Tariff frame interpolation models compatible with non-NVIDIA GPUs
New achievements and leaderboard display interface
Optimizations
Default decoding mode (without VSPipe decoding or NVIDIA fast decoding) now supports input variable frame rate videos
Other optimizations
Fixes
Fixed an issue where output resolution did not match the set resolution preset in certain cases
Fixed an issue where fast NVIDIA encoding caused audio-video desync when input was NTSC
New Features Built-in mutual help forum for posting inquiries. This also serves as an official support channel alongside the Steam forum. Software preset sharing: Choose configurations based on tags, ratings, and download count to reduce trial costs. Built-in Steam leaderboard; t
New Features
Built-in mutual help forum for posting inquiries. This also serves as an official support channel alongside the Steam forum.
Software preset sharing: Choose configurations based on tags, ratings, and download count to reduce trial costs.
Built-in Steam leaderboard; top rankings earn continuous points.
[Pro] Added a distilled version of the fast DRBA frame interpolation model.
[Pro] Turbo Mode now includes NVIDIA fast decoding and encoding options, further increasing Turbo processing speed.
[Pro] Turbo Mode adds the tariff-pwr fast frame interpolation model.
Optimizations
Optimized the process of importing a large number of files into the queue.
Improved the startup speed when processing the queue.
Improved automatic black bar detection to prevent failures in certain scenarios.
Other optimizations.
Fixes
Fixed issues where automatic cropping was incorrect in some cases.
Fixed Turbo Super Resolution errors occurring at certain resolutions.
New Features Turbo Mode supports High Precision Workflow Optimizations Turbo Mode with lower CPU usage Super Resolution Speed up Improved TensorRT compilation speed Other optimizations Fixes Fixed an issue where output videos have black blocks Other bug fixes Tutorials and Offici
New Features
[Pro] Turbo Mode supports High Precision Workflow
Optimizations
[Pro] Turbo Mode with lower CPU usage
[Pro] Super Resolution Speed up
[Pro] Improved TensorRT compilation speed
Other optimizations
Fixes
Fixed an issue where output videos have black blocks
Currently, users with 50-series NVIDIA GPUs (Blackwell architecture) need to manually switch to the beta branch Switching instructions: Select SVFI in the left software panel Right-click → Properties Navigate to Beta versions Select beta branch from dropdown and confirm Wait for
Currently, users with 50-series NVIDIA GPUs (Blackwell architecture) need to manually switch to the beta branch
Important Reminder Added new "Tariff" VFI models for NVIDIA GPUs: neu2_nb202 (fast model) and neu2_upg (quality model). The latter offers quality comparable to pg104 with over 2x faster speed! AMD users can select mnn_tariff in the algorithm dropdown (only supports nb202, slower)
Important Reminder
[Pro Version] Added new "Tariff" VFI models for NVIDIA GPUs: neu2_nb202 (fast model) and neu2_upg (quality model). The latter offers quality comparable to pg104 with over 2x faster speed! AMD users can select mnn_tariff in the algorithm dropdown (only supports nb202, slower). Recommended for 1080p resolution only; for higher resolutions, set optical flow scale below 0.5
All TensorRT engines require re-precompilation due to model naming convention changes
New Features
Added selected tasks deletion button in main interface
[Pro Version] Added GPU selection support for Turbo Mode
Added FFV1 video encoder support
Optimizations and Changes
[Pro Version] Removed upscaling ratio restrictions for image super-resolution models
[Pro Version] Removed restriction preventing concurrent use of GMFSS and super-resolution
[Pro Version] Discontinued INT8 quantization support for super-resolution
Changed deinterlacing filter from yadif to bwdif
Removed EMA VFI model
Removed Mmss VFI model
Removed 3x super-resolution models
Other optimizations
Fixes
[Pro Version] Fixed cases where super-resolution output resolution didn't match settings
Fixed playback issues for CPU-encoded HEVC videos on Apple devices
Important New Feature Introduction: Turbo Mode Significantly increases the speed of certain NVIDIA-specific models, located under Advanced Settings RIFE 4.26 VFI model accelerated by 210%+ GmfSs pg104 accelerated by 15%+ DRBA model for VFI of IRL and conservative Anime accelerate
Important New Feature Introduction: Turbo Mode
Significantly increases the speed of certain NVIDIA-specific models, located under Advanced Settings
RIFE 4.26 VFI model accelerated by 210%+
GmfSs pg104 accelerated by 15%+
DRBA model for VFI of IRL and conservative Anime accelerated by 300%+
Models under the Super Resolution TensorRT algorithm section (with .onnx extensions) receive a general speed boost
Requires purchasing the Pro version; this option may be incompatible with certain settings
New Features
[For Community Members] Supports dragging the error.log file directly into SVFI to load settings
Optimizations and Changes
The surveillance folder can now automatically detect and add newly introduced videos as new tasks, regardless of their creation time
Improved the display of vertical video posters within the software
Silent Mode now only affects the popup that appears after a task is completed. During task startup, a confirmation window will always appear to prevent unreasonable configurations
Allows the computer screen to turn off while the software is running
[Pro Version] The official_trt series VFI models have been replaced by Turbo Mode
Other optimizations
Fixes
Fixed the finishing operations after task completion so they now update in real time according to preferences
Fixed an issue where custom output filenames could sometimes cause errors
New Features Added a new feature as Surveillance folders, automatically processing tasks for video files added to target folders based on their creation time. (Hold Shift and drag a folder to quickly set it as a monitored folder) Added a 150% super-resolution preset, which may le
New Features
Added a new feature as Surveillance folders, automatically processing tasks for video files added to target folders based on their creation time. (Hold Shift and drag a folder to quickly set it as a monitored folder)
[Pro] Added a 150% super-resolution preset, which may lead to changes in some saved preset settings
Optimizations
DRBA now supports arbitrary frame rates
Prevents system sleep during task execution
Improved handling of network fluctuations during task queue execution that could cause queue interruptions
Other optimizations
Fixes
Fixed an issue where using dynamic optical flow could cause errors in some cases
Fixed an issue where using lossless encoding could cause errors in some cases
Fixed an issue where customizing output filenames could cause errors in some cases
New Features The GMFSS frame interpolation model now includes a DRBA frame interpolation mode , which only interpolates the background and not the foreground (characters), catering to users with this specific need Added the RIFE 4.20+ series to the frame interpolation models Adde
New Features
The GMFSS frame interpolation model now includes a DRBA frame interpolation mode, which only interpolates the background and not the foreground (characters), catering to users with this specific need
Added the RIFE 4.20+ series to the frame interpolation models
Added a custom output file name style option in Preferences
Added an option in Preferences to whitelist the software in Windows Defender
Added AV1 encoding support for the NVENC encoder in encoding settings
SVFI 6.18.3 Public Beta Release Compared to the official version, the content currently in public beta is as follows: Image sequence with a transparent channel and video support. Note that the image sequence needs to be saved as png or tiff, and the video will be forced to be exp
SVFI 6.18.3 Public Beta Release
Compared to the official version, the content currently in public beta is as follows:
[Professional Edition] Image sequence with a transparent channel and video support. Note that the image sequence needs to be saved as png or tiff, and the video will be forced to be exported using Apple ProRes 4444 yuva444p12le to retain the transparent channel; the speed of super-resolution and frame interpolation tasks will be halved because the transparent channel is treated as an independent video stream for processing.
[Professional Edition] NVIDIA RTX one-click HDR and fast super-resolution model. The NVIDIA APP needs to be installed.
SAIN frame interpolation model
RIFE 4.18 frame interpolation model
Problems that have been attempted to be fixed but are still in public beta for verification:
[Professional Edition] When the super-resolution ratio (e.g., 2x) does not match the model ratio (e.g., 4x), the output has a mosaic screen.
[Professional Edition] The output screen of RIFE TensorRT has jagged edges.
Interested friends, please go to the SVFI library page - Properties - Beta section, and pull down "beta" to enter the public beta.
New Features Added marker box for defining the cropping range in the player Added a search box for the super-resolution model Added AV1 support for QSVEncC encoder Added a feature to keep the SVFI window always on top to avoid potential performance degradation due to Windows 11 s
New Features
Added marker box for defining the cropping range in the player
Added a search box for the super-resolution model
Added AV1 support for QSVEncC encoder
Added a feature to keep the SVFI window always on top to avoid potential performance degradation due to Windows 11 scheduling, available in preferences
Added an automatic task restart mechanism to continue tasks after errors
Optimizations
Optimized the efficiency of regular super-resolution inference for NVIDIA 40 series graphics cards
New Features Added Interlaced input warning in the task list Added a search box for super-resolution models Added support for custom transition input (TransitionChooser) in VSPipe full workflow and image input Added support for work progress recovery using VSPipe full workflow, a
New Features
Added Interlaced input warning in the task list
Added a search box for super-resolution models
Added support for custom transition input (TransitionChooser) in VSPipe full workflow and image input
Added support for work progress recovery using VSPipe full workflow, automatic recovery only, manual recovery not supported
Added support for cropping using VSPipe decoding
Optimizations
Optimized the running efficiency of single image input
Optimized the efficiency of forcibly terminating tasks
Optimized the logic of the error message box popping up when forcibly terminating tasks
Optimized the loading logic of super-resolution models, making it easier to load custom models
Other optimizations
Fixes
Fixed an issue where the task status progress bar displayed incorrectly
Fixed an issue where audio and video were out of sync in mkv output in some cases
Fixed an issue where GPU information could not be displayed properly in some cases
Fixed support for the end moment of image input and VSPipe input
Fixed an issue where the progress display disappeared when the frame count exceeded ten thousand in VSPipe full workflow using ENCC encoding
Fixed an issue where deduplication mode dropped fewer frames than expected in some cases
Super-resolution automatic transfer resolution will be limited to full resolution to avoid severe quality degradation
New Features UI optimized Restrictions of input and output fps for TruMotion, Forward Dedup of de-duplication are removed Performance optimized Fixes Fixed an issue where ENCC output with green lines. Fixed an issue where speed down occurs in some cases. Other bug fixes.
New Features
UI optimized
Restrictions of input and output fps for TruMotion, Forward Dedup of de-duplication are removed
Performance optimized
Fixes
Fixed an issue where ENCC output with green lines.
Fixed an issue where speed down occurs in some cases.
New Features A more streamlined Advanced Settings interface. The order of options from left to right remains unchanged. To use the previous advanced settings layout, check Preferences - Classic Advanced Settings . The task input list now includes a Unknown Color Space label. This
New Features
A more streamlined Advanced Settings interface. The order of options from left to right remains unchanged. To use the previous advanced settings layout, check Preferences - Classic Advanced Settings.
The task input list now includes a Unknown Color Space label. This label indicates that the input video does not have a known color transfer characteristic to assist SVFI in correcting color shift. Manual adjustments are required in the Output Settings under Advanced Settings, specifically in Custom Color Transfer. Generally, bt709 is the default for most 1080p videos.
Fixes
Fixed an issue where the UI crashes in some cases after terminating a task.
Fixed an issue where video previews would not play correctly in some cases.
Fixed an issue with the TensorRT super-resolution module with tiling reporting errors in some cases.
Fixed an issue where constant frame rate mp4 files were merged into variable frame rate files in some cases.
Fixed an issue in multi-threading mode where it was not possible to specify the super-resolution working graphics card.
Add New Smoothness Optimization method Diff Smooth for incorrectly exported footage (like exporting an original 24fps clip to 60fps with duplicate frames). Add Scene Extrapolation to reduce duplicate frames in VFI due to scene detection. Powerful VFI models like GMFSS pg 104 are
Add New Smoothness Optimization method Diff Smooth for incorrectly exported footage (like exporting an original 24fps clip to 60fps with duplicate frames).
Add Scene Extrapolation to reduce duplicate frames in VFI due to scene detection. Powerful VFI models like GMFSS pg 104 are recommended to avoid generating artifacts.
Add new RIFE models with NCNN support.
(Pro) Add alpha(sr intensity) support for SR CUGAN TensorRT models.
