HunYuan DiT 1.0 - Open Source & Better Than Stable Diffusion 3?

Рет қаралды 12,592

22 күн бұрын

Overall, in HunYuan-DiT's tests, it scores 59.0% vs 56.7% for Stable Diffusion 3, making it objectively better. More than 50 professional evaluators performed the evaluations, so it must be true... or is it? For you to be able to judge for yourself, I pit HunYuan-DiT against the weaker (but freely available) SDXL in a battle for AI supremacy!
Want to support the channel?
/ nerdyrodent
/ hunyuan-dit-than-10582...
GitHub Repo:
github.com/Tencent/HunyuanDiT
== More Nerdy Stuff! ==
* Installing Anaconda for MS Windows Beginners - • Anaconda - Python Inst...
* Installing ComfyUI for Beginners - • How to Install ComfyUI...
* ComfyUI Workflows for Beginners - • ComfyUI Workflow Creat...
* Faster Stable Diffusions with Hyper SDXL - • Hyper-SD - Better than...
* Make A Consistent Character in ANY pose - • Reposer = Consistent S...
* Make an Animated, Talking Avatar - • Create your own animat...

Пікірлер: 78

@MrSporf 20 күн бұрын

Oh nice! We need a greater variety of models like this one. I also managed to get it running in just 6GB VRAM too - you don't need 11 now

@hakuhyo174 20 күн бұрын

Not convinced but it’s nice to have more options given recent fiasco at Stability. The game changer would be some adapter allowing new base model to use existing SD ecosystem, eg Lora, etc.

@NerdyRodent 20 күн бұрын

Fine tuning & loras are next 😊

@swannschilling474 20 күн бұрын

Thanks Nerdy! 😊

@NerdyRodent 20 күн бұрын

No problem 😊

@ritpop 21 күн бұрын

Using the negatives in Chinese isn't a problem for the SD xl version? Isn't that unfair?

@NerdyRodent 20 күн бұрын

I suggest giving it a go and running your own tests too! You’d be surprised how many other languages do things… 😉

@TamalPlays 20 күн бұрын

thanks for making video on this

@NerdyRodent 20 күн бұрын

It's my pleasure

@timothywcrane 20 күн бұрын

You said that the HunYuan couldn't do text, but it did do Hanzi (I am not sure of the correctness - would have to EN-CN OCR). Glad to see this model. Dual or more locale models coming out of China in many modals have been skyrocketing in qual and quant lately.

@TheDoomerBlox 20 күн бұрын

It does do text, and it does perform better when the English prompt is translated to Chinese by a high quality translation model first. Of course, it doesn't do ENGLISH text very well, but who's counting?

@drawmaster77 20 күн бұрын

that outro song is fire 🤣

@MilesBellas 20 күн бұрын

Fascinating comparison.

@CharlesLijt 20 күн бұрын

China No.1, thanks Rodent

@styrke9272 20 күн бұрын

Thanks for the video

@ElevatedKitten-sr6yi 20 күн бұрын

If you look closely, isn't every iteration of SD actually not true open source? Open weighs, yes - but not open source since the training data, scripts, and exact methodologies are not published. I think for this reason, SAI always carefully refers to 'open release' and similar things regarding SD3, to avoid the term 'open source' as much as possible.

@TheDoomerBlox 14 күн бұрын

The amount of webscraped Horrendous Atrocities in Picture Form present in SD1.5 would be quite a pill to swallow. :- ) and SD3 still responds to webscraped metadata tags, so it's still there

@TobyDeshane 20 күн бұрын

Should probably be using stock SDXL to completely avoid wondering if a problem is with SDXL itself or not. :P (Even if those models do generate often better imagery in general.) Additionally, while I don't know what your actual workflow is before editing the video, I'd probably want to generate 3-4 images for a prompt (or maybe cherry picking the best out of 6 or more?), since it's quite common for a generated image to miss some aspect of the prompt. I know this isn't a strictly scientific review, but if we're going to be pitting A vs B even in a fluff intro video like this, it's not really doing anyone favors for either one of them to be at the mercy of a random number generator. (Maybe you do this behind the scenes and just edit that bit out -- that's fine, but I can't tell that from the end result! :D) Anyway, cheers -- interesting seeing alternative models coming out.

@generalawareness101 20 күн бұрын

I am sure you are the type that wants it in an X/Y grid too. Am I right?

@suliao-lv6gj 20 күн бұрын

Love your rigor and skepticism, but it's open source and you can totally test it once for yourself. I hope you keep that same rigor and skepticism when you see other videos and ask for proof yourself.

@lex_darlog_fun 20 күн бұрын

Dear Nerdy Rodent, could you please publish a link to your udio-generated outro? I'd like to extend it to a full track.

@NerdyRodent 20 күн бұрын

I may do a short one day 😉

@lex_darlog_fun 20 күн бұрын

@@NerdyRodent Hope this day comes sooner than later. I was hooked by the minimalistic yet vibing melody since your udio review.

@rsunghun 21 күн бұрын

Nice

@neonskimmer 18 күн бұрын

That closing credits song reminds me a lot of a British band from the early 2000s called Fat Truckers. You'd like them I think.

@sandy66555 20 күн бұрын

Lots of handsome rodents, and a bunny? What happened to my badger? You promised (you didn't but I took it that way) 😜

@NerdyRodent 20 күн бұрын

The badger is currently chilling 🦡

@DeathMasterofhell15 17 күн бұрын

problem mate my loader does not have vae or clip or model but instead "pipeline, auto encoder, scheduler ? how to change it

@NerdyRodent 17 күн бұрын

The HunYuan pipeline loader is detailed at github.com/Tencent/HunyuanDiT/tree/main/comfyui-hydit#hunyuan-pipeline-loader

@soloones7141 20 күн бұрын

When I try to queue prompt, it gives me this error "DiffusersPipelineLoader: - Value not in list: pipeline_folder_name: 'ckpts' not in [ ]". Any idea why, Mr. Nerdy Rodent?

@pon1 15 күн бұрын

Don't know for sure, when I had a similar problem it was because I had my models at a different directory than the regular directory (I set the path of the models to where I have the A111 models), maybe it's expecting something else there or is expecting the ComfyUI default models path. I ended up installing ComfyUI at a different drive to use with models like SD3 (which is the model that I had a similar problem with).

@popecillo 20 күн бұрын

Definitely, review the models you use, they showed that they are too biased, there are too many mixes of models that are very ruined in their understanding, the base model contains ample knowledge but it is lost the more they train it without a good balance, now the mixes Most models put everything inside and there are models that improve certain concepts but ruin others. In my tests using your prompts I didn't have problems with women's bias or terribly deformed hands.

@generalawareness101 20 күн бұрын

the SD1.5 plague has arrived to SDXL, and it plays hell on we who train loras for it. Might work on yours, or might not, or do weird stuff. Just too many mixes, of mixes that was trained on a mix type fiasco.

@Afr0man4peace 20 күн бұрын

It would be interesting to compare it with a actual good trained model like Colossus Project XL 10 for example.

@zappazack 11 күн бұрын

Hunyuan seems not to work on windows as a build wheel error occurs

@CMak3r 15 күн бұрын

Actually this one might be better than SD3 that we got. I should invest in bigger GPU

@CMak3r 20 күн бұрын

According to steam hardware survey, about 1% has 4090 and 0.58% have 3090

@lefourbe5596 20 күн бұрын

132 million active user on Steam. 1% of steam gamers have a 4090 that would be 1.23 million peoples. More than the view this video will every have. (I have a 3090 btw)

@ArawnFR 20 күн бұрын

Your maths are wrong, 2.3% better than sd 3 (56.7) is 58.0041, not 59

@gimperita3035 15 күн бұрын

The song is a hit 😂

@NerdyRodent 15 күн бұрын

😎

@Mp3Pintyo 20 күн бұрын

I also tested it in a long video, but it really didn't convince me. kzfaq.info/get/bejne/ZtFkdch8s9XRo6c.html

@MilitantHitchhiker 19 күн бұрын

Better than the unreleased weights of SD3? Huh amazing that they can compare their model with a model that isn't finished training and can definitively tell theirs is better. What a time to be alive, technology is marvelous.

@leslieviljoen 14 күн бұрын

In what world is Dall-e better than Midjourney?

@cowlevelcrypto2346 5 күн бұрын

Wait, I can't run ComfyUI in Linux?

@NerdyRodent 5 күн бұрын

Other way around - ComfyUI runs best in Linux 😉

@KlimovArtem1 20 күн бұрын

Can it do any nsfw, or it's heavy censored?

@joechip4822 20 күн бұрын

Whether it does NSFW ist the first question any reviewer should ask and answer. Considering this is Chinese, the answer will be obvious though, won't it?

@KlimovArtem1 20 күн бұрын

@@joechip4822 the level of censorship in any particular model is one of the most important aspects.

@lefourbe5596 20 күн бұрын

It's a non problem. Sure it's an arm cut off the integrity but talented fine tunner are able to add that back seamlessly.

@suliao-lv6gj 20 күн бұрын

While my answer may not be entirely accurate, it's subject to Chinese law and NSFW content is prohibited.

@joechip4822 20 күн бұрын

@@suliao-lv6gj very probably true - but also a good reason not to use and support it in its original form

@dayantargaryen4129 15 күн бұрын

it's much better

@financialfreedomfighters369 20 күн бұрын

According to this chart, HunYuan is supposed to be almost as good as Midjourney in terms of aesthetics and Dalle-3 is supposed to be even better than Midjourney... The curator of this chart cleary needs glasses x-D

@openroomxyz 20 күн бұрын

Well can someone create safetensor versions of these ? xD

@generalawareness101 20 күн бұрын

do it yourself as the tools are out there in kohya with lycoris.

@TamalPlays 20 күн бұрын

sadly it is only 1024x1024 :(

@v0idbyt3 17 күн бұрын

time to put my rtx 3060 12gb + 64gb ddr4-2400 ubuntu pc to use

@Smirnoff67 20 күн бұрын

who would have guessed, a model that doesnt obey to some self restriction obtain better result

@nunuarthas8680 20 күн бұрын

sir it's pronounced Hung-Yu-Whan, hunyuan

@NerdyRodent 20 күн бұрын

Thank you! I’m not good with anything other than British pronunciations 😅

@li_tsz_fung 15 күн бұрын

If that's supposed to be a Chinese word, @nerdyrodent's pronunciation is closer. But the name makes no sense to me anyway.

@jairuskersey8311 12 күн бұрын

Asking for a friend, can Hunyuan do NSFW stuff?

@taucalm 21 күн бұрын

SD3 aint released yet so how can he compare something that doesnt even exist yet?

@NerdyRodent 20 күн бұрын

I imagine they used the api 😉

@j5545 20 күн бұрын

It doesn't exists? Lol

@TamalPlays 20 күн бұрын

sd3 api has been released

@taucalm 20 күн бұрын

@@j5545 "Stable Diffusion 3 (SD3) was announced by Stability AI as their most advanced text-to-image model to date. The early preview of SD3 was made available in February 2024, with the general release of a more accessible version, known as SD3 Medium, scheduled for June 12, 2024. This model aims to improve photorealism and prompt adherence, making it suitable for both consumer and business applications (Stability AI) (THE DECODER) (Decrypt)." Where I live is now 9. of june. So it doesnt exist for me yet. You must come from future.

@JoernR 12 күн бұрын

One major drawback - UncomfyUI. :/

@generalawareness101 20 күн бұрын

this thing is too bloated and my my comfy is set up to work in bfloat while this is set up for float so mat1 and mat2 errors. When I go to 32bit this model is slow. I mean my 6,5it/s is now 2s/it. Not very good quality either.

@SimosFunk 20 күн бұрын

@NerdyRodent 20 күн бұрын

😃

@denisquarte7177 3 күн бұрын

nope, sd3 has better architecture but safety training as well as license are dogsh*. And you know these scores can be gamed. No way in hell dalle3 beat midjourney. I use all of them on a daily basis.