**A collection of StarSample V2.0 models, now trained in a few architectures, rather than just one.**

The dataset for V2 is made completely from scratch (in relation to V1, but also in general), and includes what I'd consider many smaller datasets purpose-made to target areas where the models couldn't quite get up to the quality I wanted (particularly with focus on the heavier-weight models — the lighter architectures aren't really big enough for much effect). The improvement in dataset mostly, but also training heavier models has lead to a very large improvement, particularly with the "HQ" model.

I wouldn't suggest using V1 over V2 in any circumstances — which I am very pleased to be able to say, having gone through at least 20 attempts to actually achieve that... somehow.

StarSample V2.0

This is a model for the restoration of My Little Pony: Friendship is Magic, however it also works decently well on similar art.

V2.0 greatly improves upon V1.0's dataset in every way, taking models from (realistically) only being viable at 1x, to now being far more competent at 2x, more so for the models trained with heavier architectures in this release. 

Improvements come as a significantly better understanding of compressions, and partly architecturally/partly dataset improved handling of details and overall understanding of content, leading to less artifacting and "AI smudging". The dataset takes from a larger variety of sources, despite being smaller than V1.0 (when tiled V1.0 would be 71,876 pairs), due to being filtered for IQA scores and detail density. It also contains many thousands of image pairs manually created to cover areas where there wasn't sufficient information.

This release also includes "NS", or "No Scale" models, which are a better representation of my initial goal with StarSample, and (StarSample V2.0 NS) should provide great 1x restoration results with little apparent artifacting, even where the heavier 2x models can fail due to having to increase resolution.

- 2x StarSample V2.0 HQ — _(HAT-L)_
- 2x StarSample V2.0 — _(ESRGAN)_
- 2x StarSample V2.0 Lite — _(SPAN-S)_
- 1x StarSample V2.0 NS — _(ESRGAN)_
- 1x StarSample V2.0 Lite NS — _(SPAN-S)_ — **THIS MODEL**

[Github Release](https://github.com/Derpiesaurus/models/releases/tag/v2.0_Lite_NS)

StarSample V2.0 Lite NS

This is a model for the restoration of My Little Pony: Friendship is Magic, however it also works decently well on similar art.

V2.0 greatly improves upon V1.0's dataset in every way, taking models from (realistically) only being viable at 1x, to now being far more competent at 2x, more so for the models trained with heavier architectures in this release. 

Improvements come as a significantly better understanding of compressions, and partly architecturally/partly dataset improved handling of details and overall understanding of content, leading to less artifacting and "AI smudging". The dataset takes from a larger variety of sources, despite being smaller than V1.0 (when tiled V1.0 would be 71,876 pairs), due to being filtered for IQA scores and detail density. It also contains many thousands of image pairs manually created to cover areas where there wasn't sufficient information.

This release also includes "NS", or "No Scale" models, which are a better representation of my initial goal with StarSample, and (StarSample V2.0 NS) should provide great 1x restoration results with little apparent artifacting, even where the heavier 2x models can fail due to having to increase resolution.

- 2x StarSample V2.0 HQ — _(HAT-L)_
- 2x StarSample V2.0 — _(ESRGAN)_
- 2x StarSample V2.0 Lite — _(SPAN-S)_
- 1x StarSample V2.0 NS — _(ESRGAN)_ — **THIS MODEL**
- 1x StarSample V2.0 Lite NS — _(SPAN-S)_

[Github Release](https://github.com/Derpiesaurus/models/releases/tag/v2.0_NS)

StarSample V2.0 NS

Purpose: Anime

This is a successor to AnimeSharpV3 based on RCAN instead of ESRGAN. It outperforms both versions of AnimeSharpV3 in every capacity. It's sharper, retains *even more* detail, and has very few artifacts. It is __extremely faithful__ to the input image, even with heavily compressed inputs. 

To use this model, you must update to the [**latest chaiNNer nightly build**](https://github.com/chaiNNer-org/chaiNNer-nightly/releases)

The `2x-AnimeSharpV4_Fast_RCAN_PU` model is trained on RCAN PixelUnshuffle. This is much faster, but comes at the cost of quality. I believe the model is ~95% the quality of the full V4 RCAN model, but ~6x faster in Pytorch and ~4x faster in TensorRT. This model is ideal for video processing, and as such was trained to handle MPEG2 & H264 compression.

__Comparisons:__ https://slow.pics/c/63Qu8HTN
https://slow.pics/c/DBJPDJM9
![1736292155 679079](https://github.com/user-attachments/assets/875be696-0948-4de2-9d03-85a581891a49)

AnimeSharpV4

AnimeSharpV4_Fast_RCAN_PU

[Link to Github Release with more infos to the process](https://github.com/Phhofm/models/releases/tag/2xAoMR_mosr)

2xAoMR_mosr  
Scale: 4  
Architecture: [MoSR](https://github.com/umzi2/MoSR)  
Architecture Option: [mosr](https://github.com/umzi2/MoSR/blob/95c5bf73cca014493fe952c2fbc0bdbe593da08f/neosr/archs/mosr_arch.py#L117)  

Author: Philip Hofmann  
License: CC-BY-0.4  
Purpose: A 2x mosr upscaling model for game textures  
Subject: Game Textures  
Input Type: Images  
Release Date: 21.09.2024 (dd/mm/yy)  

Dataset: Game Textures from Age of Mythology: Retold  
Dataset Size: 13'847  
OTF (on the fly augmentations): No  
Pretrained Model: [4xNomos2_hq_mosr](https://github.com/Phhofm/models/releases/tag/4xNomos2_hq_mosr)  
Iterations: 510'000  
Batch Size: 4  
Patch Size: 64  

## Description:   
In short: A 2x game texture mosr upscaling model, trained on and for (but not limited to) Age of Mythology: Retold textures.

Since I have been playing Age of Mythology: Retold (casual player), I thought it would be interesting to train an single image super resolution model on (and for) game textures of AoMR, but this model should be usable for other game textures aswell.
This is a 2x model, since the biggest texture images are already 4096x4096, I thought going 4x on those would be overkill (also there are already 4x game texture upscaling models, so this model can be used for similiar cases where 4x is not needed).

Model Showcase:
[Slowpics](https://slow.pics/s/IlkmsToH)

2xAoMR_mosr

Mixed 3D/2D CGI animations on some 2D animations it might sharpen the edges. General usage is to upscale CGI animations compressed by YouTube.

CGIMaster v1

This model is designed to upscale the standard definition digital era of the Pokémon anime, which runs from late season 5 (Master Quest) to early season 12 (Galactic Battles). During this time, the show was animated digitally in a 4:3 ratio. This process was also used for Mewtwo Returns, most of Pokémon Chronicles, and the Mystery Dungeon specials.

Advice/Known Limitations:
* This OmniSR model can occasionally produce black frames when run in fp16 mode. This seems to be more common in the TPCi era (seasons 9 and later). The issue is sporadic enough that it probably makes sense to do a first pass in fp16, then re-upscale any affected shots in fp32.
* I recommend using QTGMC on a preset of "Slow" or slower for deinterlacing. While the show is primarily animated at 12/24 fps, some elements like backgrounds are animated at a full 60i, particularly for the Diamond & Pearl era, which increased the base animation frame rate to 15 fps.
* The model is not great at handling fonts, particularly the italicized text in the episode credits. This is despite including font images in the training data.

Version History:

* 1.1.0 (9/16/2025): Added images from SD portion of season 12, augmented training to better handle slightly resized images.
* 1.0.0: Initial Release

Digital Pokémon-Large

This model is designed to upscale the standard definition digital era of the Pokémon anime, which runs from late season 5 (Master Quest) to early season 12 (Galactic Battles). During this time, the show was animated digitally in a 4:3 ratio. This process was also used for Mewtwo Returns, most of Pokémon Chronicles, and the Mystery Dungeon specials.

Advice/Known Limitations:
* I recommend using QTGMC on a preset of "Slow" or slower for deinterlacing. While the show is primarily animated at 12/24 fps, some elements like backgrounds are animated at a full 60i, particularly for the Diamond & Pearl era, which increased the base animation frame rate to 15 fps.
* The model is not great at handling fonts, particularly the italicized text in the episode credits. This is despite including font images in the training data.

Version History:

* 1.1.0 (9/16/2025): Added images from SD portion of season 12, augmented training to better handle slightly resized images.
* 1.0.0: Initial Release

Digital Pokémon-Small

Purpose: 2x anime upscaler which tries to preserve dof

2x anime upscaler, a non-otf version of the HFA2kCompact model which tries to preserve dof by not deblurring.

Original Link: https://drive.google.com/drive/folders/1RZbALUTCf4DYDTVMBR74b8Nt-7m6TEyk?usp=drive_link

HFA2k compact multijpg

Upscale Anime while keeping as much of the original detail as possible. Isn't among the sharpest models and if the anime is somewhat blurry it will retain that detail but add more pixels to give a smoother edge. Doesn't change the color very much and may make some scene's very slightly darker.

Pretrained model: Pony Models

Loyaldk Giroro

Loyaldk Keroro

Loyaldk Kororo

Upscale MLP episodes. Liter version. Able to handle compression better compared to V1.0 and no longer creates halo's/rainbows. Good for Vector 2D art however converts detail to blobs.

Loyaldk LitePony V2.0

Upscale MLP episodes. Liter version. Able to handle compression better compared to V1.0 and no longer creates halo's/rainbows. Good for Vector 2D art however converts detail to blobs. Leaves more blobs when converting detail to blobs compared to litepony

Loyaldk MediumPony V2.0

Upscale MLP episodes. Full version. Able to handle compression better compared to V1.0 and no longer creates halo's/rainbows. Good for Vector 2D art however converts detail to blobs.

Loyaldk SuperPony V2.0

This is a model for the restoration of My Little Pony: Friendship is Magic, however it also works decently well on all similar art. 

It was trained in 2x on ground truth 3840x2160 HRs and 1920x1080 LRs of varying compression, so it is able to upscale from 1080p to 2160p, where its detail retention is great, however it may create noticeable artifacting if looked at closely, like areas of randomly coloured pixels along edges. In 1x or 1.5x (2x upscaled and then downscaled back down) it performs extremely well, almost perfectly in fact, in correcting colours, removing compression, and crisping up lines - and this is the way the model is intended to be used (hence the acronym of its name being "SS", or "supersampling").

**[Github Release](https://github.com/Derpiesaurus/models/releases/tag/v1.0)**   
**Showcase:** https://slow.pics/s/1ixqCSjy

StarSample V1.0

Purpose: Compression Removal, Noise Reduction, Line Correction, MPEG2 / LD Artifact Removal

This is my first model release. The model was inspired from a personal project which I have been pursuing for some time now, which this model aims to solve. This model will upscale low resolution hand drawn animation from 1970s to 2000. The colors are retained with effective noise control. Details and Textures are maintained to a good degree considering animations. Color Spills are also corrected depending on the colors. Shades of white and yellows have been difficult. It also makes the lines slightly sharper and thinner. This could be a plus depending on your source. The model is also temporally stable across my tests with little observable issues. 

**Showcase:** 
Images - https://imgsli.com/MjYwNzY1/12/13
Video Sample -  https://t.ly/Jp7-w vs Upscale - https://t.ly/PdsKs

2x Pooh V4

This is a model for the restoration of My Little Pony: Friendship is Magic, however it also works decently well on similar art.

V2.0 greatly improves upon V1.0's dataset in every way, taking models from (realistically) only being viable at 1x, to now being far more competent at 2x, more so for the models trained with heavier architectures in this release. 

Improvements come as a significantly better understanding of compressions, and partly architecturally/partly dataset improved handling of details and overall understanding of content, leading to less artifacting and "AI smudging". The dataset takes from a larger variety of sources, despite being smaller than V1.0 (when tiled V1.0 would be 71,876 pairs), due to being filtered for IQA scores and detail density. It also contains many thousands of image pairs manually created to cover areas where there wasn't sufficient information.

This release also includes "NS", or "No Scale" models, which are a better representation of my initial goal with StarSample, and (StarSample V2.0 NS) should provide great 1x restoration results with little apparent artifacting, even where the heavier 2x models can fail due to having to increase resolution.

- 2x StarSample V2.0 HQ — _(HAT-L)_
- 2x StarSample V2.0 — _(ESRGAN)_ — **THIS MODEL**
- 2x StarSample V2.0 Lite — _(SPAN-S)_
- 1x StarSample V2.0 NS — _(ESRGAN)_
- 1x StarSample V2.0 Lite NS — _(SPAN-S)_

[Github Release](https://github.com/Derpiesaurus/models/releases/tag/v2.0)

This is a model for the restoration of My Little Pony: Friendship is Magic, however it also works decently well on similar art.

V2.0 greatly improves upon V1.0's dataset in every way, taking models from (realistically) only being viable at 1x, to now being far more competent at 2x, more so for the models trained with heavier architectures in this release. 

Improvements come as a significantly better understanding of compressions, and partly architecturally/partly dataset improved handling of details and overall understanding of content, leading to less artifacting and "AI smudging". The dataset takes from a larger variety of sources, despite being smaller than V1.0 (when tiled V1.0 would be 71,876 pairs), due to being filtered for IQA scores and detail density. It also contains many thousands of image pairs manually created to cover areas where there wasn't sufficient information.

This release also includes "NS", or "No Scale" models, which are a better representation of my initial goal with StarSample, and (StarSample V2.0 NS) should provide great 1x restoration results with little apparent artifacting, even where the heavier 2x models can fail due to having to increase resolution.

- 2x StarSample V2.0 HQ — _(HAT-L)_ — **THIS MODEL**
- 2x StarSample V2.0 — _(ESRGAN)_
- 2x StarSample V2.0 Lite — _(SPAN-S)_
- 1x StarSample V2.0 NS — _(ESRGAN)_
- 1x StarSample V2.0 Lite NS — _(SPAN-S)_

[Github Release](https://github.com/Derpiesaurus/models/releases/tag/v2.0_HQ)

StarSample V2.0 HQ

This is a model for the restoration of My Little Pony: Friendship is Magic, however it also works decently well on similar art.

V2.0 greatly improves upon V1.0's dataset in every way, taking models from (realistically) only being viable at 1x, to now being far more competent at 2x, more so for the models trained with heavier architectures in this release. 

Improvements come as a significantly better understanding of compressions, and partly architecturally/partly dataset improved handling of details and overall understanding of content, leading to less artifacting and "AI smudging". The dataset takes from a larger variety of sources, despite being smaller than V1.0 (when tiled V1.0 would be 71,876 pairs), due to being filtered for IQA scores and detail density. It also contains many thousands of image pairs manually created to cover areas where there wasn't sufficient information.

This release also includes "NS", or "No Scale" models, which are a better representation of my initial goal with StarSample, and (StarSample V2.0 NS) should provide great 1x restoration results with little apparent artifacting, even where the heavier 2x models can fail due to having to increase resolution.

- 2x StarSample V2.0 HQ — _(HAT-L)_
- 2x StarSample V2.0 — _(ESRGAN)_
- 2x StarSample V2.0 Lite — _(SPAN-S)_ — **THIS MODEL**
- 1x StarSample V2.0 NS — _(ESRGAN)_
- 1x StarSample V2.0 Lite NS — _(SPAN-S)_

[Github Release](https://github.com/Derpiesaurus/models/releases/tag/v2.0_Lite)

StarSample V2.0 Lite

Purpose: Animation, Pixel Art

This model has VERY strong compression removal and line restructuring that allows it to restore any heavily compressed drawings, animation, cartoons, or anime. Also works on games as well as DDS compression. It renders frames very quickly and is very viable for restoring videos.

Pretrained model: 4x-MMScale (Yes, 4x. mistakes were made)

UniScale CartoonRestore Lite

Upscaling model for 1990s cartoon series "Where on Earth is Carmen Sandiego?" May also be useful for other shows of that era.

Note: Before using this model, you will need to either:
* Deinterlace the episodes (I recommend QTGMC on Slower) to 60 fps (I recommend 60 fps because that's the rate of the live-action footage and some other components).
* Repair the existing deinterlacing via one of the related models for this show.
Which method will depend on the episode and release. See the related deinterlacing models for details.

This show is...complicated. Unlike, say, the Super Mario Bros. Super Show, where live action and animated segments are generally separated into large chunks, Where on Earth is Carmen Sandiego repeatedly switches between live-action, traditional animation, 3D CGI, 2D computer graphics, stock photography, and other sources, sometimes combining them in the same frame. As such, rather than create multiple models for each type, I've created a single model for all footage types.

Known Weaknesses:
* The model performs less well on the live-action (Player) and 3D CGI (C5 Corridor) segments due to a scarcity of usable data in those scenes.
* I don't own the 2025 DVD release, so I can't comment on how well this model works on that footage.

Where On Earth-Upscale-Large

Where On Earth-Upscale-Small

[Real-ESRGAN](https://github.com/xinntao/Real-ESRGAN) aims at developing Practical Algorithms for General Image/Video Restoration.

Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.
We extend the powerful ESRGAN to a practical restoration application (namely, Real-ESRGAN), which is trained with pure synthetic data.

RealESRGAN_x2Plus

A 2x model designed for hand-drawn animation of the 80's, 90's and early 2000s.

The Super Mario Bros. Super Show is a hybrid animated and live-action show that debuted in 1989. Each episode featured a cartoon segment sandwiched between live-action segments. The combination of the two formats makes the show very difficult to upscale, and creates the need for multiple model files.

These models were trained using footage from the show itself (except for some title images that were recreated in Photoshop) by finding cases where the same image was shown at different resolutions. For the animated footage, I also did some retouching to reduce haloing, color bleed, and other issues.

A few notes:

- The opening sequence (in particular, the first shot with the giant talking Mario face) is very difficult to upscale due to the picture being degraded from multiple layers of compositing. Ideally, a professional remaster of the series would have access to the individual elements, upscale those, then recomposite them.
- Although live-action models are provided, you will likely get better results from commercial upscaling software (I recommend  the Proteus model in Topaz Video AI).
- Footage must be deinterlaced before processing (I recommend QTGMC for both live-action and animated segments).

[Full Read Me](https://www.mediafire.com/file/brp4c0pvv5s42rw/ReadMe.txt/file)

[Sample Animated Result](https://www.mediafire.com/view/wtieecra7iov35f/Animation_Sample.png/file)

[Sample Live Action Result](https://www.mediafire.com/view/xa7jenxl18xd1y7/Live_Action_Sample.png/file)

Architecture	SPAN
Scale	2x
Size	48nf
Color Mode	RGB
License	CC-BY-NC-SA-4.0 Private use Distribution Modifications Credit required Same License State Changes No Liability & Warranty Disclaimer
Date	2026-02-12
Dataset	HR = 4K GT uncompressed MLP: FiM episode frames + relevant uncompressed HR pairs to LR datasets /// LR = 1080p MLP: FiM episode frames sourced from YouTube in 3 different bitrates + custom MLP: FiM focal blur dataset + custom MLP: FiM GIF compression dataset in 3 different compression levels + custom MLP: FiM difficult details and other edge cases dataset + custom artificially-degraded MLP: FiM background dataset
Dataset size	53560
Training iterations	149
Training epochs	500000
Training batch size	16
Training HR size	192
Training OTF	No

StarSample V2.0 Lite

Collections that include this model

Similar Models