DataVoid @DataPlusEngine Twitter profile

Pinned Tweet

DataVoid

@DataPlusEngine

1 month

Dreaming... #AIart

1

4

34

Last Seen Profiles

@CiriacoRonaldo

@BinorRaja

@NSAS1011

@TheZack_Zilla

@Boothy6

@KyleXaka

@stw_pdg

@uludagdev

@ESpitzRadio

@manuelturizogt

@Tristian2025

@Nabil00Aziz

@RottazSA

@IMGAcademy

@SirHenry_

@citrusshrtcake

@lilmeme555

@VmaniakJ

@stw_pdg

@tkg00_Q

@xi_phracanthr

@Da___Engineer

@JackermanDev

@zuhra380

@stw_pdg

@sgyemikale

@AndrewDuffEU

@g_onthebeach

@bokeplokalmalam

@instant2d

@ensdommains

@FeodaleHorige

@LightwolfM

@bagdaeung281802

@MillieDreams101

@GPSA_CU

DataVoid

@DataPlusEngine

10 months

Have Fun <3! I merged 5 models together into something close to dalle3 level by accident! SDXL DPO is the main helper in achieving this!

dataautogpt3/OpenDalle · Hugging Face

huggingface.co

22

46

366

DataVoid

@DataPlusEngine

7 months

𝙒𝙝𝙤 𝙣𝙚𝙚𝙙𝙨 𝙪𝙥𝙨𝙘𝙖𝙡𝙚𝙧𝙨? The TempestV0.1 Initiative is a powerhouse in image generation, leveraging an unparalleled dataset of over 6 million images. The collection's vast scale, with resolutions from 1400x2100 to 4800x7200, encompasses 200GB of high-quality

34

29

180

DataVoid

@DataPlusEngine

5 months

SDXL ELLA is expected to be achieved internally by the end of the day today.

20

8

168

DataVoid

@DataPlusEngine

5 months

i just finished what i think was the last finetune.... I am not one to say this by my god this might be dangerous to open source! Hands down MJV6 level. this is not a joke or exaggeration i feel at this point.

Ippi

@Coolzippity

5 months

The new proteus mobius model has: 1. The most diverse outputs 2. The strongest prompt control 3. The best aesthetic Of any model I've used to date. I mean just look at the perspective on these!

1

2

47

13

8

141

DataVoid

@DataPlusEngine

3 months

Civitai bans all use of SD3 related resources and content on their website

Temporary Stable Diffusion 3 Ban | Civitai

Unfortunately, due to a lack of clarity in the license associated with Stable Diffusion 3 , we are temporarily banning: All SD3 based models All mo...

civitai.com

27

25

110

DataVoid

@DataPlusEngine

5 months

I have joined @Corcel_X as Head of Research! I will be working full time as a Applied ML Researcher/Head of their diffusion Research department. working on open sourcing state of the art Diffusion based models.

Corcel.io

@Corcel_X

5 months

Thrilled to unveil that @DataPlusEngine has come aboard the Corcel Research Labs team, spearheading our Image Generation research. Our mission: to craft the premier open source image generation model. Stay tuned for groundbreaking updates in the weeks ahead! 😍

2

15

95

19

12

108

DataVoid

@DataPlusEngine

4 months

This flew under everyone's radar!

GitHub - comfyanonymous/ComfyUI_TensorRT

Contribute to comfyanonymous/ComfyUI_TensorRT development by creating an account on GitHub.

github.com

3

9

102

DataVoid

@DataPlusEngine

3 months

@multimodalart

dataautogpt3/PixArt-Sigma-900M · Hugging Face

huggingface.co

3

12

98

DataVoid

@DataPlusEngine

1 month

dataautogpt3/FluxteusV1 · Hugging Face

huggingface.co

6

19

99

DataVoid

@DataPlusEngine

1 month

FLUX-aestheticAnime

2

6

96

DataVoid

@DataPlusEngine

1 month

dataautogpt3/FLUX-MonochromeManga · Hugging Face

huggingface.co

5

15

94

DataVoid

@DataPlusEngine

4 months

I implemented me own "prompt injection", Now I can... prompt SPECIFIC BLOCKS of the UNet in different ways, And I implemented three different kinds of nodes for different workflow usages Each of these on the screenshot does the same thing, inject a specific prompt in OUTPUT0

7

16

93

DataVoid

@DataPlusEngine

9 months

I am happy to present and announce ! a fine-tune of my OpenDalleV1.1 using 220k captioned gptV dataset to align and then dpo tuned on 10k dalle3 captioned image pairs. results are another step above OpenDalle in terms of prompt following and style.

6

13

88

DataVoid

@DataPlusEngine

8 months

… after some serious consideration and thought. I just open sourced both ProteusV0.2 and ProteusV0.1 for Commercial use under GLP 3.0.

dataautogpt3/ProteusV0.2 · Hugging Face

huggingface.co

6

19

82

DataVoid

@DataPlusEngine

6 months

We reverse engineered the ELLA training for 1.5 and successfully made a finetune of it. We are working on adapting the script to work with SDXL. major disappointment in them for not releasing it. So when in doubt do it yourself!

9

6

83

DataVoid

@DataPlusEngine

1 month

plastic render looking issue fixed? FLUX-detail

8

0

76

DataVoid

@DataPlusEngine

4 months

SD3 Support. · comfyanonymous/ComfyUI@8c4a9be

github.com

5

8

76

DataVoid

@DataPlusEngine

2 months

34B for those saying flux was too small

6

5

75

DataVoid

@DataPlusEngine

1 month

FLUX-MonochromeManga (stack of 14 different loras)

5

3

72

DataVoid

@DataPlusEngine

4 months

This cycle has repeated every release. Step 1. Angry Anticipation: "WHERE THE F*CK IS THE MODEL!" Step 2. Release Rage: "THIS MODEL F*CKING SUCKS" Step 3. Reversion Resignation "I'm going back to 1.4, it make t*ddies good" Step 4. Furries F*ck is released "This model is so

15

3

71

DataVoid

@DataPlusEngine

3 months

The PixArt Sigma 900M really helps fix the issue of grainy faces that the 600M base had! still undertrained right now. going to talk and cover information about it and Prometheus with @HelloCivitai tomorrow during their podcast/interview with me!

2

5

71

DataVoid

@DataPlusEngine

1 month

Found way to handle the Loras onto the base checkpoint without causing a leaking on styles/concepts across each other. Fluxteus is born!

6

5

71

DataVoid

@DataPlusEngine

5 months

In other news, I just finished Retraining SD1.5 for complete scratch! examples are not upscaled, more prompt adherent than SDXL for sure.

9

7

71

DataVoid

@DataPlusEngine

3 months

Trained SDXL arch from scratch. Still very much undertrained atm. This is NOT associated with Corcel, This is my own personal pet project.

7

4

71

DataVoid

@DataPlusEngine

2 months

Okay, I am now convinced a large amount of the OpenAI fanboys have actually lost their marbles now

grey

@greynewell

2 months

the world is yours

51

12

181

4

0

70

DataVoid

@DataPlusEngine

3 months

trained a 900M pixart Sigma model as apposed to the 600M base. here are some samples.

7

4

70

DataVoid

@DataPlusEngine

2 months

PrometheusV2 is SOTA. I am frankly convinced of that based on my testing over the last hour.

5

4

68

DataVoid

@DataPlusEngine

2 months

@Yampeleg here! ✋

DataVoid

@DataPlusEngine

2 months

@gospaceport here is mine! 16X3090s

10

2

48

7

0

67

DataVoid

@DataPlusEngine

3 months

The 900M Pixart Sigma Model should be done cooking by this upcoming Wednesday. shout out to for making his adaptation to the base model possible. (it produces noised images since its not trained)

jimmycarter (Jimmy Carter)

huggingface.co

1

3

65

DataVoid

@DataPlusEngine

2 months

Just achieved full rank finetuning of Stable Diffusion 3 Medium 2B on a SINGLE consumer GPU! Only months ago, this was thought possible only on 80GB VRAM cards. Now: ✅ Model: SD3 2B ✅ Hardware: 24GB VRAM RTX 4090 ✅ Dataset: First layer of Proteus (prompt-alignment subset)

4

7

65

DataVoid

@DataPlusEngine

2 months

Flux-syntheticanime lora at 100% strength.

7

5

64

DataVoid

@DataPlusEngine

2 months

Found the issue that was causing the lack of detail. retrained the FLUX-syntheticanime lora overnight and its a lot better. going to redo megalith next. the quality is so so so so much better now.

5

4

62

DataVoid

@DataPlusEngine

4 months

ITS LIVEEE!

stabilityai/stable-diffusion-3-medium · Hugging Face

huggingface.co

1

11

61

DataVoid

@DataPlusEngine

3 months

yo WTF, even Edward Snowden is even shitting on SD3?!?!

Edward Snowden

@Snowden

3 months

Appalling that @StabilityAI so ruthlessly censored SD3 (for fear that someone might draw—gasp!—a nipple) that it nearly lost the ability to draw women entirely. Behold, "A woman lying in the grass:" Fire the entire "AI Safety" team and replace them with an anatomy textbook.

313

638

4K

12

4

60

DataVoid

@DataPlusEngine

3 months

sorry for the delay regarding the 900M Pixart Sigma model. Been extremely sick, never got hit with covid all the way up until now. went to the docs yesterday because it got so bad. they have me on quick special dial because i also have a autoimmune condition.

22

0

61

DataVoid

@DataPlusEngine

3 months

this is a MASSVIVE deal

Ostris

@ostrisai

3 months

I trained a new VAE with 16x depth and 42 channels (kl-f16-d42). I am now training SD1.5 to work with it, which will double the output size of SD1.5 without much additional compute overhead. Every time I train a new latent space, it always starts out inverted. It's so odd.

14

34

235

4

5

58

DataVoid

@DataPlusEngine

3 months

900M PixArt Sigma - base | PixArt sigma Checkpoint | Civitai

PixArt Sigma 900M: Enhanced Text-to-Image Model Model Overview PixArt Sigma 900M is a text-to-image generation model based on the PixArt Sigma arch...

civitai.com

4

7

59

DataVoid

@DataPlusEngine

2 months

@HEAVY_WASHH GET MONKEYED

0

58

DataVoid

@DataPlusEngine

6 months

lady's and gents we are about to work on the models and bot full time with 3 employees under my company DataPusle AI. this is all thanks to investment by a undisclosed party (will be revealed in a bit)

13

2

58

DataVoid

@DataPlusEngine

2 months

Novel flux-dev conditioning method in ComfyUI: Independently condition CLIP and T5-XXL with architecture-specific prompts to influence style and content. Example: T5-XXL (content): "a cat sitting on a bench drinking a latte in anime style." CLIP (style): "cat, anime, anime

DataVoid

@DataPlusEngine

2 months

found a cool way to condition flux-dev in a way that you can squeeze actual styles out of it! it messes with text tho

1

0

37

6

9

55

DataVoid

@DataPlusEngine

3 months

900M Pixart Sigma is almost done cooking. again it fixes to a decent degree the issue the previous model had with faces. (its not quite done yet)

2

1

55

DataVoid

@DataPlusEngine

2 months

Oh, also, btw I got Flux training working First Lora will be a synthetic anime lora because I absolutely love the style! Btw, I found a way to even get a 32 rank lora to train on 24vram. instead of just 16 rank. I'll make a pull request

3

2

53

DataVoid

@DataPlusEngine

2 months

Does anyone want to collaborate on making a FLUX lora? I have some very, very well curated datasets. I have datasets ranging in size from 400K to just 120. The Flux training has been very finicky however. Please let me know!

11

0

53

DataVoid

@DataPlusEngine

5 months

Finished the retune of ProteusV0.3

6

1

51

DataVoid

@DataPlusEngine

5 months

new arch just finished training on over 1.5 million images. results are looking promising. any questions are welcome!

9

2

53

DataVoid

@DataPlusEngine

2 months

🚀 Double release day! 🎉 1️⃣ PrometheusV1: • First full rank finetune of Playground v2.5 • Optimized for open-source accessibility • Custom CLIP integration (use clip skip 2!) 2️⃣ ProteusV0.5 is out too! Pushing boundaries in #AIArt . Try them

dataautogpt3/PrometheusV1 · Hugging Face

huggingface.co

2

9

52

DataVoid

@DataPlusEngine

6 months

Here is our working implementation of Training ELLA 1.5! still working on getting it working for SDXL!

GitHub - DataCTE/ELLA_Training

Contribute to DataCTE/ELLA_Training development by creating an account on GitHub.

github.com

Sébastien Carbain

@SCarbain

6 months

@DataPlusEngine That's great, congrats ! Can you already share the training script for 1.5 🙏 ?

0

2

7

52

DataVoid

@DataPlusEngine

5 months

today is my bday frens! send hugs! 🤗

24

0

51

DataVoid

@DataPlusEngine

2 months

📣 ProteusV0.5 Release: 🧠 Custom CLIP 🖼️ 400k+ dataset: Enhanced photorealism & style diversity ⚙️ Optimized LORA integration for targeted model improvements HF: Civitai:

Proteus - v0.5 | Stable Diffusion Checkpoint | Civitai

Proteus Proteus serves as a sophisticated enhancement over OpenDalleV1.1, leveraging its core functionalities to deliver superior outcomes. Key are...

civitai.com

2

11

51

DataVoid

@DataPlusEngine

4 months

We gave the @FAL early access to the upcoming Mobius model and its only been up on for 3 hours. its already the best stable diffusion based image model in the world based on human preference data

Ian

@IanDikhtiar

4 months

@DataPlusEngine It's insane, my dude. Clear winner most of the times. If this thing is open sourced - you wll clearly be the best open source model provider on the market.

0

1

4

8

51

DataVoid

@DataPlusEngine

4 months

Our upcoming paper outlines and enables making entirely new base diffusion models without the need to extensively pretrain a new model from scratch. We can in a controlled way, break all the quality and composition associations without damaging the baseline understanding from the

5

2

49

DataVoid

@DataPlusEngine

4 months

(mobiusV1.1) cooking again

6

3

48

DataVoid

@DataPlusEngine

5 months

4

3

49

DataVoid

@DataPlusEngine

4 months

MobiusV1.3 has been achieved internally. still improving the model before release. improved anime and detailing overall without any compromise

8

2

49

DataVoid

@DataPlusEngine

4 months

my friend Mobi who helped created the prompt injection node just made a clip switch node for SDXL clip to SD3! no weird requirements.txt, super light, extra simple, but make sure to read the README of the repo,

GitHub - G-370/sd3-clip-swap: Ever wanted to yoinkysploinky SDXL's clip and use them in SD3? Well,...

Ever wanted to yoinkysploinky SDXL's clip and use them in SD3? Well, now you can! - G-370/sd3-clip-swap

github.com

5

8

49

DataVoid

@DataPlusEngine

2 months

@gospaceport here is mine! 16X3090s

10

2

48

DataVoid

@DataPlusEngine

4 months

(Mobius)

2

48

DataVoid

@DataPlusEngine

4 months

Pro tip for using the Mobius model on @huggingface : For realistic photos, set the CFG to 3.5. For artistic images, crank it up to 7.0. Experiment with these settings to get the best results! cfg 3.5 image below!

4

5

47

DataVoid

@DataPlusEngine

2 months

PrometheusV2.7 is live on my Discord bot! please help me test it out! (its arguable sota in terms of aesthetics and comp)

3

5

47

DataVoid

@DataPlusEngine

6 months

working on getting a implementation of #SD3 working with a sponsor. We are fairly confident that we can get one trained. wish us luck! more announcements coming soon.

9

3

47

DataVoid

@DataPlusEngine

4 months

it also has the most range of any model ever made aside from MJ.

6

5

46

DataVoid

@DataPlusEngine

6 months

@StabilityAI

0

1

42

DataVoid

@DataPlusEngine

3 months

thank you to @multimodalart for setting up this space for me!

PixArt Sigma 900M - a Hugging Face Space by dataautogpt3

huggingface.co

0

6

46

DataVoid

@DataPlusEngine

5 months

samples from the current continued training run are looking great. 1B params, it should run on a potato. also MJ data was not used period. throw some prompts at me!

14

4

45

DataVoid

@DataPlusEngine

2 months

Announcing Black Forest Labs

Today, we are excited to announce the launch of Black Forest Labs. Deeply rooted in the generative AI research community, our mission is to develop and advance state-of-the-art generative deep…

blackforestlabs.ai

5

8

45

DataVoid

@DataPlusEngine

4 months

reddit is a cesspit my god

19

2

44

DataVoid

@DataPlusEngine

2 months

I put a lot of effort into making the playground 2.5 arch/model backwards compatible with as many existing SDXL tools, loras, samplers, and etc. as possible! that's one of the key aspects of the model!

Matteo Spinelli

@cubiq

2 months

Prometheus from the good friend @DataPlusEngine seems to be working with IPAdapter which is interesting considering it's a Playground finetune.

0

4

43

2

7

45

DataVoid

@DataPlusEngine

4 months

Corcel.io

@Corcel_X

4 months

Big News! We're so excited to unveil Corcel Research Labs' two new image models: Open Vision & Mobius! ✨ Both fully open sourced under the apache-2.0 licence! Check them out here: 💖Open Vision: Finetuned from S19's MJ

25

45

227

3

44

DataVoid

@DataPlusEngine

6 months

We Will be Training SDXL 3B model from scratch aswell as a MMDIT (SD3) 15B model from scratch (sample from our sdxl finetune)

5

4

43

DataVoid

@DataPlusEngine

4 months

playing around with a custom sampler that has very fine control and i am loving it! choose the noise map and everything custom selection of dpm or eular per step two phase sampling etc

4

43

DataVoid

@DataPlusEngine

3 months

Finished the first Finetune over night!

6

3

43

DataVoid

@DataPlusEngine

7 months

ProteusV0.4: The Style Update This update enhances stylistic capabilities, similar to Midjourney's approach, rather than advancing prompt comprehension. Methods used do not infringe on any copyrighted material.

6

9

41

DataVoid

@DataPlusEngine

2 months

flux-megalith lora 100% strength.

4

0

42

DataVoid

@DataPlusEngine

4 months

@josemwarrior

7

1

40

DataVoid

@DataPlusEngine

4 months

Pretty sure this is the FINAL version of Mobius internal before release. look at the details!

2

41

DataVoid

@DataPlusEngine

2 months

DataVoid

@DataPlusEngine

2 months

and here is what i assume people actually want, PrometheusV2-aesthetic. I would recommend using the refiner with it. also it has a trained trigger word of "aesthetic".

7

1

18

6

5

40

DataVoid

@DataPlusEngine

1 month

5

6

40

DataVoid

@DataPlusEngine

5 months

finally after months of work i found a way to apply my techniques to 1.5! all of these images have less than 5 word prompts!

3

1

41

DataVoid

@DataPlusEngine

7 months

@qtnx_ i am rather agnostic regarding timelines and tend to lean on the quote "When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong"

In 1903, New York Times predicted that airplanes would take 10 million years to develop

Only nine weeks later, the Wright Brothers achieved manned flight. The pathologically cynical always will find a reason to complain.

bigthink.com

1

39

DataVoid

@DataPlusEngine

4 months

4

40

DataVoid

@DataPlusEngine

7 months

@Locati0ns i wish i was a fish rn

1

0

39

DataVoid

@DataPlusEngine

3 months

Prometheus is clearing up!

4

0

39

DataVoid

@DataPlusEngine

7 months

You can also selectively target what the model is allowed to be applied to within the input :)

6

5

38

DataVoid

@DataPlusEngine

3 months

(Prometheus) what's what i am going to call the SDXL tuned from scratch

2

1

39

DataVoid

@DataPlusEngine

9 months

Exciting News! OpenDalle v1.1 is here! We've revved up the realism & style, staying true to your prompts. Think of it as a step above SDXL, yet playfully chasing DALLE-3's tail. Get it now 👉

6

7

37

DataVoid

@DataPlusEngine

4 months

🚀 Exciting news from Corcel! We're previewing our upcoming image models and need your feedback to fine-tune it with DPO. Your input will help make it better for everyone. Check it out now and let us know what you think: #AI #MachineLearning #Feedback

Corcel

Corcel API

corcel.io

3

8

37

DataVoid

@DataPlusEngine

4 months

mobi friend who helped me make the prompt injection is doing it again ;)

GitHub - G-370/ComfyUI-SD3-Powerlab

Contribute to G-370/ComfyUI-SD3-Powerlab development by creating an account on GitHub.

github.com

2

3

38

DataVoid

@DataPlusEngine

4 months

from testing aswell shift completely fixes the comp issues. from 6 to 10. y'all need to chill out its been out for less than a day lmao. its like driving a new car you have to give it some time to get used to the new system. chill yourself out yall lolol

4

2

38

DataVoid

@DataPlusEngine

7 months

#SD3 better watch out, i still have a few cards up my sleeve. sample output:

4

2

36

DataVoid

@DataPlusEngine

5 months

have text working consistently with Mobius(changing the name from ProteusMobius to just Mobius)

7

2

37

DataVoid

@DataPlusEngine

7 months

𝗣𝗿𝗼𝘁𝗲𝘂𝘀-𝗥𝘂𝗻𝗗𝗶𝗳𝗳𝘂𝘀𝗶𝗼𝗻'𝘀 "𝗦𝘁𝘆𝗹𝗲 𝗨𝗻𝗹𝗼𝗰𝗸𝗶𝗻𝗴" 𝗶𝗻 𝗮𝗰𝘁𝗶𝗼𝗻: We applied our innovative CLIP retraining approach to Animagine XL V3, an anime-only model, with remarkable results! Without any retraining of the model itself, our novel CLIP

RunDiffusion.com

@RunDiffusion

7 months

Introducing Proteus v0.4 + RunDiffusion Photo 'Style Unlocking' 🚀🎨: A game-changer in AI art generation. Our retrained CLIP model breaks through stylistic boundaries, unlocking endless creative possibilities. This has the potential to unlock numerous SDXL models! #AIart #ai

5

8

43

6

4

37

DataVoid

@DataPlusEngine

2 months

3

2

37

DataVoid

@DataPlusEngine

8 months

ProteusV0.3: The Anime Update

4

7

36

DataVoid

@DataPlusEngine

4 months

higher cfg fixes the autonomy issues it seems

Emad

@EMostaque

4 months

Some tips for stable diffusion 3 #SD3 Sampler: dpmpp_2m (not karras etc) Steps: 32 CFG: 4-6

12

32

307

8

2

36

DataVoid

@DataPlusEngine

5 months

might open source this model under a very restrictive license. thoughts?

14

1

36

DataVoid

@DataPlusEngine

2 months

found a cool way to condition flux-dev in a way that you can squeeze actual styles out of it! it messes with text tho

1

0

37

DataVoid

@DataPlusEngine

3 months

what if we make SDXL..... but bigger!

5

1

37

DataVoid

@DataPlusEngine

1 month

also FLUX_photoreal is done i think

5

1

37

DataVoid

@DataPlusEngine

3 months

trained a Anime Variant along with the 900M base. So there will be two models 900M-base and a 900M-anime! prompts: left: a woman smiling right: a epic swords man

0

4

36

DataVoid

@DataPlusEngine

1 month

FLUX-manga, it actually has consistent characters across panels. made with @FAL . @isidentical lol ani fusion should use this

4

1

37