A free, open-source Windows player that runs your videos, photos and GIFs through NVIDIA's DLSS 5 neural renderer on your own RTX card while you watch, keeps the original one key away, on the same frame, and saves the result as a file.
Windows · NVIDIA RTX · driver 610.47+ · MIT licensed. A community project, not an NVIDIA product. Is it safe?
Drag the seam above. Both sides are one frame of the 007 First Light story trailer: the source on the left, the player's render at its default settings on the right, cropped identically and not retouched, sharpened or colour-graded. The change sits in skin, stubble and light across the face: the renderer rebuilds detail rather than laying a filter over the picture.
Source2560×1440 · 59.94 fps · 96.5 s
Renderthe whole video, at the default settings, upscaling off
MachineRTX 4080 SUPER · driver 610.47
007 First Light and Grand Theft Auto VI at the default settings: a paused frame split down the face, GTA VI playing with the divider held still, then the player's Difference, Side by side and loupe views, and a live render filling the timeline (shown at 2× speed). No sound.
Rendered on an RTX 4080 SUPER; how it was made. 007 First Light — IO Interactive; Grand Theft Auto VI — Rockstar Games.
Each still is one frame of a trailer, the source on the left and the player's render on the right, the same unscaled 700×880 crop at the default settings. They are the strongest frames found, not typical ones; the frame where it does not help is with the limits.
Flip swaps the picture between the two; 1:1 shows the captured files at one image pixel per screen pixel.
Rendering is not an export queue. Press D at any point and playback moves onto rendered frames a few seconds later, while the rest of the video renders behind you, nearest the playhead first. Seek anywhere: rendered frames play wherever they are, and elsewhere the original plays at once while the render moves there.
Split, wipe, side by side or 2 × 2, all on one timestamp, from a compare bar under the picture. Drag the divider across a face, hold the mouse to see the original in place, or step frames on either view.
Difference shows the change itself, amplified 1× to 32×. The loupe magnifies both at the same spot, and a setting changed while paused re-renders that frame, so you decide on the picture.
Wherever the window shows a video larger than it is, NVIDIA’s RTX Video Super Resolution makes the picture at that size, on by default. Compare against RTX VSR (Shift+R) puts it beside the plain scale on the same frame. In a separate measurement, six clips from 960×540 to 1080p, it scored 87.5 VMAF against bicubic’s 80.5 (report).
Save comparison image writes exactly what the picture shows as a PNG, with a footer recording the player version, video, frame, view, neural settings and runtime.
Renders stay on disk until you clear them or the drive falls below 20 GB free. Reopen a film weeks later and its whole-film render is still attached, as long as the source, the runtime and the neural settings still match. From 0.28.1, an update keeps them too, unless its notes say otherwise.
Open a PNG, JPEG, BMP, TIFF or WebP and press D. The same model renders it, the same compare views show what changed, and it saves back as a PNG or JPEG at full size. Here the game’s smooth orange skin comes back with texture, a truer tone and light that falls across the face. Animated GIFs render frame by frame and keep their timing.
A video, a photo or an animated GIF: whatever you watch rendered, you can save. Save converted video writes the render you already have without rendering again; Export with DLSS stages (Ctrl+S) renders a new file. Both are under DLSS > Convert & export.
VideoMKV or MP4, of the whole video or a marked clip. MKV keeps the source audio, subtitles and chapters without re-encoding them.
GIFAn animated GIF renders frame by frame with its own timing, and saves as GIF, MP4 or MKV.
PhotoPNG, JPEG, BMP, TIFF or static WebP in; PNG or JPEG out, at the source size. The 2560×1440 screenshot above took 8 s from the command line on an RTX 4080 SUPER.
StagesSuper Resolution, neural rendering and frame generation at 2× to 5× the frame rate, any combination, in one file and in NVIDIA’s order. Super Resolution on its own runs on RTX VSR by default, or DLSS Super Resolution if you choose it.
Scripteddlss5-convert clip.mp4 writes clip-dlss.mkv without opening the player, and takes folders and wildcards too. Every stage and Neural setting is an option: --preset strong --passes 3 stacks three passes of the strong look, --stages fg --multiplier 3 triples the frame rate. dlss5-convert probe says what a file is and what this machine can run on it; the usage guide has the rest.
Left outSubtitles travel as tracks and are never burned in. Image adjustments and playback upscaling are not baked in; the stages export is how a larger size goes into the file.
Details are in the usage guide. Nothing is uploaded: the work happens on your GPU and the cache stays on your disk.
dlss5-convert converts files and folders, skips what is done and reports each resultgh attestation verifyverify_package.ps1 checks every file against a pinned manifest before the first run
Stated as plainly as the features, and measured where a number is given.
HardwareAn NVIDIA RTX card and driver 610.47 or newer; older drivers are refused up front. Checked on an RTX 4080 SUPER and an RTX 5090. RTX 20 and 30 cards lack native FP8, should be several times slower, and are untested; for their frame generation the player offers the unofficial community dlssg_sm86 add-on.
SpeedOn an RTX 5090 a live session costs 6.8 ms a frame at 1080p30, 9.9 ms at 1440p30 and 23 ms at 4K30, all faster than the video plays. On an RTX 4080 SUPER a whole 4K render ran at 14.8 fps, slower than a 30 fps film plays.
UpscalingRTX Video Super Resolution upscales playback by default: 87.5 VMAF against 80.5 for bicubic on six clips, 960×540 to 1080p. DLSS Super Resolution scored below bicubic on all six; it is built for games, so it is off by default.
RuntimeThe neural runtime is a modified, unsigned community build, not a signed NVIDIA package. Why it is unsigned.
ArtifactsDepth is estimated from the picture, and so is motion on cards without the optical flow engine. Expect some artifacts.
Low keyIn deep shadow the model can crush a face towards black and add speckle. The frame below is the plainest case found.
CacheA render started partway into a video is not joined to the rest, and below 20 GB free the least recently watched renders are deleted first.
Several open-source projects put DLSS 5’s neural pass on video. This one is built to watch the result and check it: it renders the whole video while you watch, keeps the render, and shows the original beside it on the same frame. As of 2 October 2026, from each project’s own README, releases and issues; the full list has more, and corrections are welcome.
| This player | Visual Enhancer v14.0 | NeuralScreen v2.1.9 | Veyra 2.0.0 | DLSS5Tool v2.3.3 | |
|---|---|---|---|---|---|
| What it is | Plays the video while rendering all of it in the background, and exports | Converts through a stage pipeline; a Live mode plays through it | An overlay on the whole desktop, plus file conversion | Plays files, capture cards and game streams in real time | Converts, with a preview |
| Render you keep | Yes: rendered frames are cached, seekable and re-watched | No: Live keeps a 2–30 s buffer | No: real time | No: real time | Export only |
| Compare | Split, Wipe, Difference, Side by side, 2×2, loupe, hold for original, against RTX VSR | Split, 2-Up | Before/after wipe | Original/enhanced toggle | Wipe, side by side |
| Upscaling | RTX VSR by default in playback and export; DLSS SR on request; measured | RTX VSR and DLSS SR stages, off by default | None, by design | DLSS SR, RTX VSR or FSR | RTX VSR 2× or 4× |
| Frame gen. | DLSS-G 2–5× as a conversion; RTX 20/30 through an offered, hash-checked add-on | DLSS-G as a conversion | DLSS-G 2–4×, live | DLSS-G, XeSS or FSR up to 6×, live | DLSS-G 2×, export |
| Inputs | Local video, public YouTube, photos, GIFs | Images, video, URLs, YouTube, Twitch | Anything on screen, files | Files, images, capture cards, PS5, PC and Xbox streams | Images, video, image sequences |
| Export keeps | Audio, subtitles and chapters | Audio and chapters; not subtitles | Audio | Audio and subtitles | Audio |
| HDR | Shown on HDR displays; exports are SDR | 10-bit HDR export; SDR→HDR | HDR10 recording; capture experimental | Kept; HEVC Main10 export | HDR10/HLG export |
| Command line | dlss5-convert: files, folders, JSON report | VE_CLI.exe | No; a GUI queue | Not documented | Not documented |
| Download | 377 MB, or 36 MB without the runtime | 726 MB | 230 MB | 409 MB | 580 MB lite |
| Verify | SHA-256, GitHub attestations, a manifest and a verifier | GitHub’s per-file digest | SHA256SUMS and a runtime manifest | SHA-256 files | Runtime hashes |
| Licence | MIT | Source-available, its own terms | PolyForm Strict, non-commercial | GPL-3.0; AGPL for streaming | MIT |
Of the five, only this player keeps the render, so a seek lands on rendered frames and a video you come back to is already done. It compares the render with the original on the same frame in more ways than any of the others, RTX VSR included; only its export keeps subtitles and chapters as well as audio; and only its downloads carry GitHub attestations of what this repository built.
None of these projects is an NVIDIA product, and each runs the neural pass on a runtime NVIDIA has not published for this use, this one included. This player is published only on its releases page; re-uploads of DLSS 5 tools under other accounts have been reported carrying malware.
v0.28.2, published 4 October 2026. Free, with no account and no installer: unzip it and run it. Every build is on the release page with its checksum.
The player and the pinned neural runtime. This is the one you want.
Download 377 MBdlss5-video-player-v0.28.2-win64.zip
The player alone, without the neural runtime. Build or supply your own.
Download 36 MBDLSSVideoPlayer-v0.28.2-core-win64.zip
MIT licensed. Read it, build it yourself, or check that the core zip was built from it.
View source 164 starsgithub.com/2600th/dlss5-video-player
neural-runtime/ next to the executable.DLSSVideoPlayer.exe.
It is open source, but its neural runtime is a modified community
build that nobody can sign, so decide with that in view. Each zip has a
SHA-256 beside it. The core zip is built by CI from this repository and
carries a signed build attestation (gh attestation verify);
the complete zip is assembled by the maintainer and has a checksum
only. Inside either, verify_package.ps1 checks every file
against a pinned manifest before the first run. Scan the runtime as you
would any unsigned software, or take the core zip and supply your own.
Because it has been modified. It includes NVIDIA's neural DLL modified by the community to run on every RTX generation, 20 through 50, which invalidates NVIDIA's signature, alongside community ReShade and RenoDX files that were never signed. Nobody is in a position to re-sign it, so it ships unsigned and says so. Its files are pinned by size and SHA-256, and the package verifier rejects any that have changed.
An NVIDIA RTX card and driver 610.47 or newer; with an older driver the player says so up front. It has been checked on an RTX 4080 SUPER and an RTX 5090. RTX 20 and 30 series cards lack native FP8 and should be several times slower; nobody has tested neural rendering on one yet. On any other GPU the start screen says "not an RTX GPU, so no DLSS", and playback there has not been tested. Frame generation is NVIDIA's for RTX 40 and 50; on RTX 20 and 30 the player offers to download the unofficial community dlssg_sm86 add-on, and a user ran frame generation at 4x on an RTX 3060 with it.
You do not wait for it to start: playback joins the render a few seconds after you press D, and the rest fills in while you watch. On an RTX 5090 a live session costs 6.8 ms a frame at 1080p30, 9.9 ms at 1440p30 and 23 ms at 4K30, all faster than the video plays. An RTX 4080 SUPER measured 15.3 ms at 1080p30 on an older version, and rendered a whole 4K film at 14.8 frames a second. After your first session the player warns you before a video it expects to fall behind on. The other limits are above.
No. Rendering runs on your GPU and the cache stays on your disk. The
player goes online to fetch a YouTube link you paste, and once a day
to ask GitHub whether there is a new release, which
[Updates] Enabled=0 in its settings file turns off.
The start screen also fetches the seven built-in trailers' thumbnails
from YouTube's image server, then again once a month, which
[Start] ThumbnailFetch=0 turns off.
Yes. RTX Video Super Resolution upscales playback by default wherever the window shows a video larger than it is: on six clips upscaled from 960×540 to 1080p it scored 87.5 VMAF against 80.5 for a plain bicubic upscale. DLSS Super Resolution is there too, off by default, because on video it scored below bicubic every time. Neural rendering itself keeps the source resolution.
Yes. Open a PNG, JPEG, BMP, TIFF or static WebP and press D: it renders once and stays on that frame, with the same compare views as a video, and saves as a PNG or JPEG at its own size. From the command line, a 2560×1440 game screenshot became a PNG in about 8 seconds on an RTX 4080 SUPER. Animated GIFs render frame by frame and keep their timing. The model is built for rendered games, and game screenshots are what it has been tried on; camera RAW and animated WebP are not supported.
Yes: PNG or JPEG for photos, GIF for animations, MP4 or MKV for video,
from the player or from the command line with
dlss5-convert, which also converts whole folders. MKV keeps the source audio, subtitles and
chapters without re-encoding them.
Export with DLSS stages writes Super Resolution, neural rendering and
frame generation into one file, in NVIDIA's order.