Search
514 results for “db0”
-
If I can’t dance, it’s not my revolution
I see tankies keep trying to argue with people about "Actually Existing Socialist" states like USSR and China and try to argue with me or others about how "they were good actually". It's bad enough when most of their arguments are whataboutism, but it grinds my gears when I hear then prattle on about all the statistically significant material improvements the life of the people received. It's like listening to a terminally-liberal prattle on about how "statistically, the life quality […]https://dbzer0.com/blog/if-i-cant-dance-its-not-my-revolution/
-
[Video] El Fin de La Policia
Los levantamientos por la muerte de George Floyd en los Estados Unidos en 2020 despiertaran el debate sobre la abolición de la policía. ¿Pero es posible vivir sin la policía? ¿Para qué fue creada la policía y qué papel juega en el mundo de hoy?
https://kolektiva.media/videos/watch/10f1f228-fbba-4d3f-afb0-db036bd600dc
#acab #abolicióndelapolicía #usa #GeorgeFloyd #viendo #peertube
-
Die neue Staffel #Kroymann ist so witzig. Bin bei Folge 2. @Maren_Kroymann -
I wish #firefish best luck, but I doubt moving to a different DB is going to be the end of their performance issues. I'm honestly surprised this is the approach they went for, instead of just tuning postgresql as it's an incredibly mature piece of software.
Always remember, never, ever host any serious #database on a VM/VPS. Always go for #baremetal if you have the choice.
I've had to learn this #TheHardWay multiple times...
-
"Twitterrific was our flagship app, and by far earned the most of any of our pieces of software we’ve produced. Its loss is still being felt revenue-wise to this day. We're hopeful that Tapestry will help us regain some of that ground and help us keep our company healthy, doing the things we love to do — developing great, friendly and useful apps for iOS and macOS."
The linked article though is quite interesting in what it reveals from all of this. When you consider both Twitter and Reddit did this to their users and 3rd party devs, there is a trend towards these more private centralised social networks to have restrictions in the way of 3rd party access. Whilst on the decentralised social networks the trend tends to be using an open protocol, and usually also supporting open RSS feeds.
So, Tapestry is tapping into this more open ecosystem: "This isn't really a social media client directly in the mold of Twitterrific; instead, it's a single place to view online stuff that matters, and it mostly uses open standards like ActivityPub and RSS. Luckily for them, they've already reached their $100,000 minimum funding goal."
Its intention is not to replace any existing clients, although with all the decentralised social networks I've been involved with, their devs all openly encourage having different 3rd party client apps.
But it's what is not directly stated in the linked article that concerns me the most, and I've heard others also echoing this sentiment, especially some journalists who have "jumped ship" from Twitter to Mastodon and other networks. The issue is around quality content creation, especially news. It's not that there is no such news on the decentralised social networks, it's that there are still too many mainstream news services, journalists, and governments stuck fast on Twitter, Reddit, etc. So many users are also just stuck in their bubble in the same places.
We do really need to see that final seismic shift across to embracing an open and decentralised social network ecosystem. Certainly, the Fediverse, built on the ActivityPub protocol, is not the only option out there, but it is an open standard endorsed social network and has gained a lot of content and user traction.
The way I understand Tapestry, is that it would not only help provide an integrated (single) view of the Fediverse (yes most existing ActivityPub apps can also do that), but it can also seamlessly include discovering and following content from Bluesky, and many others, including potentially even Nostr (via RSS feeds).
So, Tapestry may help provide some cohesion and single view that many users and content creators, who are still stuck on Twitter, Reddit, etc, seem to crave/demand. Many such users are actually caught up in indecision over "where must I pick to move to". With something like Tapestry, that fear may disappear, and for creators such as news and governments they would be more easily "found and followed" through one single app like Tapestry.
Will it happen? I think so as the vast majority of users are often solely content consumers, and whilst it is true that to develop a high-quality app from the ground up takes cash, Iconfactory does have that experience and they have raised over US$100,000 to fund their app development.
Who knows, I'd expect this app to even go further, and start to support likes and comments for some of the networks it integrates with over time.
See After Elon Gutted Their App, Iconfactory Wants To Make A New Billionaire-Proof Social Network? | DiggIconfactory made delightful Twitter clients for Mac and iOS for over a decade, but they've had to pivot to something much bigger since a certain billionaire got in the way.
#technology #socialnetworks #Twitterrific #Tapestry -
Noch etwas näher am QTH am #Raichberg in der Nähe von #DB0RAB mit Blick auf #Hohenzollern.
-
Film the Police DC: If you see something, say something.
https://social.trom.tf/display/dbc8dc44-1768-b842-bea9-db0776116993
#policestate #WashingtonDC #DC #OccupiedDC #authoritarianism #fascism #resist #news #resource #action
-
Having a long discussion with one of the more reasonable MLs from #hexbear, and it's downright fascinating https://lemmy.dbzer0.com/post/29900244/14179136
I'm really pushing them to argue the "left unity" which they claim exists in hexbear between anarchists and MLs and the best they can come up with is that hexbear anarchists accept "Actually Existing Socialism" as a fact (which is just...wut?!) and that they all hang out in the same place doing their own leftist things.
-
Nach ihrer Einstufung als "gesichert rechtsextremistisch" will die AfD gegen den Verfassungsschutz klagen. Wie stehen die Chancen?#AfD #Verfassungsschutz #Einstufung #gesichertrechtsextremistisch #Jurist #Gericht
AfD will gegen Einstufung klagen: Hat die Partei Chancen vor Gericht? -
The good part is that my recognized value with my #FOSS work, finally has hardened enough within me, that it keeps the self-doubt away. So when I get rejected from giving a conference talk about AI (as is what happened with #fosdem2024 ) or having my right to belong to something like #Nivenly constantly challenged, I can remind, nay, *prove* to myself that I am actually worth it, but the system has always been rigged against neurodivergent people from the start.
-
The good part is that my recognized value with my #FOSS work, finally has hardened enough within me, that it keeps the self-doubt away. So when I get rejected from giving a conference talk about AI (as is what happened with #fosdem2024 ) or having my right to belong to something like #Nivenly constantly challenged, I can remind, nay, *prove* to myself that I am actually worth it, but the system has always been rigged against neurodivergent people from the start.
-
The good part is that my recognized value with my #FOSS work, finally has hardened enough within me, that it keeps the self-doubt away. So when I get rejected from giving a conference talk about AI (as is what happened with #fosdem2024 ) or having my right to belong to something like #Nivenly constantly challenged, I can remind, nay, *prove* to myself that I am actually worth it, but the system has always been rigged against neurodivergent people from the start.
-
The good part is that my recognized value with my #FOSS work, finally has hardened enough within me, that it keeps the self-doubt away. So when I get rejected from giving a conference talk about AI (as is what happened with #fosdem2024 ) or having my right to belong to something like #Nivenly constantly challenged, I can remind, nay, *prove* to myself that I am actually worth it, but the system has always been rigged against neurodivergent people from the start.
-
The good part is that my recognized value with my #FOSS work, finally has hardened enough within me, that it keeps the self-doubt away. So when I get rejected from giving a conference talk about AI (as is what happened with #fosdem2024 ) or having my right to belong to something like #Nivenly constantly challenged, I can remind, nay, *prove* to myself that I am actually worth it, but the system has always been rigged against neurodivergent people from the start.
-
In an online address to union members Wednesday, union President Shawn Fain said General Motors, Ford and Stellantis have raised initial wage offers, ...#uawnegotiations2023update #uawstrike #uawstrikeupdate #uawstrike2023 #uaw #unitedautoworkers #stellantis #uawnegotiations #gmstrike #fordstrikefordstrike2023 #uawcontractnegotiations
UAW update: United Auto Workers, automakers remain far from a deal as strike deadline approaches -
The initial deployment of the Stable Cascade (SC) on the AI Horde supported just text2image workflows, but that was just a subset of what this model can do. We still needed to onboard the rest of its capabilities.
One such capability was the “image variations” option, which allows you to send an image to the model, and get a variation of that image, perhaps with extra stuff added in, using the unClip technology. This required quite a bit of work on hordelib so that it uses a completely different ComfyUI workflow but ultimately this was not so much harder than just adding the img2img capabilities to SC.
The larger difficulty came when I wanted to add the feature to remix multiple images together. The problem being that until now the AI Horde only supported sending a single source image and a single source mask, so a varying amount of images was not possible at all.
So to support this, I needed to touch all areas of the AI Horde. The AI Horde had to accept and upload each of them on my R2 bucket and provide individual download links. The SDK had to know to expect and provide methods to download those images in parallel to avoid delays, to the reGen worker had to be able to receive those images and send them to hordelib which should know how to dynamically adjust a comfyUI pipeline on-the-fly to add as many extra nodes as required.
So after 2 weeks of developing and testing, we finally have this feature available. If your Horde front-end supports the “remix” feature. You can send up to 1-6 images to this workflow along with a prompt, and it will try its best to “squash” them all together into one composition. Note that the more images you send, and the larger the prompt, the harder it will be for the model to “retain” all of them in the composition. But it will try its best.
As an example, here’s how the model remixes my own avatar. You’ll notice that the result can understand the general concepts of the image, but can’t follow it exactly as it’s not doing img2img. The blur is probably caused by the need to upscale my original image, which is something I’d like to fix on the next pass.
Original AvatarRemix AvatarLikewise, this is the Haidra logo
Original LogoRemix LogoAnd finally, here’s a remix of both logo and avatar together
Pretty neat, huh?
This ability to send extra source images also lays the groundwork for the Horde to support things like InstantID, which I hope I’ll be able to work on supporting soon enough.
-
The initial deployment of the Stable Cascade (SC) on the AI Horde supported just text2image workflows, but that was just a subset of what this model can do. We still needed to onboard the rest of its capabilities.
One such capability was the “image variations” option, which allows you to send an image to the model, and get a variation of that image, perhaps with extra stuff added in, using the unClip technology. This required quite a bit of work on hordelib so that it uses a completely different ComfyUI workflow but ultimately this was not so much harder than just adding the img2img capabilities to SC.
The larger difficulty came when I wanted to add the feature to remix multiple images together. The problem being that until now the AI Horde only supported sending a single source image and a single source mask, so a varying amount of images was not possible at all.
So to support this, I needed to touch all areas of the AI Horde. The AI Horde had to accept and upload each of them on my R2 bucket and provide individual download links. The SDK had to know to expect and provide methods to download those images in parallel to avoid delays, to the reGen worker had to be able to receive those images and send them to hordelib which should know how to dynamically adjust a comfyUI pipeline on-the-fly to add as many extra nodes as required.
So after 2 weeks of developing and testing, we finally have this feature available. If your Horde front-end supports the “remix” feature. You can send up to 1-6 images to this workflow along with a prompt, and it will try its best to “squash” them all together into one composition. Note that the more images you send, and the larger the prompt, the harder it will be for the model to “retain” all of them in the composition. But it will try its best.
As an example, here’s how the model remixes my own avatar. You’ll notice that the result can understand the general concepts of the image, but can’t follow it exactly as it’s not doing img2img. The blur is probably caused by the need to upscale my original image, which is something I’d like to fix on the next pass.
Original AvatarRemix AvatarLikewise, this is the Haidra logo
Original LogoRemix LogoAnd finally, here’s a remix of both logo and avatar together
Pretty neat, huh?
This ability to send extra source images also lays the groundwork for the Horde to support things like InstantID, which I hope I’ll be able to work on supporting soon enough.
-
The initial deployment of the Stable Cascade (SC) on the AI Horde supported just text2image workflows, but that was just a subset of what this model can do. We still needed to onboard the rest of its capabilities.
One such capability was the “image variations” option, which allows you to send an image to the model, and get a variation of that image, perhaps with extra stuff added in, using the unClip technology. This required quite a bit of work on hordelib so that it uses a completely different ComfyUI workflow but ultimately this was not so much harder than just adding the img2img capabilities to SC.
The larger difficulty came when I wanted to add the feature to remix multiple images together. The problem being that until now the AI Horde only supported sending a single source image and a single source mask, so a varying amount of images was not possible at all.
So to support this, I needed to touch all areas of the AI Horde. The AI Horde had to accept and upload each of them on my R2 bucket and provide individual download links. The SDK had to know to expect and provide methods to download those images in parallel to avoid delays, to the reGen worker had to be able to receive those images and send them to hordelib which should know how to dynamically adjust a comfyUI pipeline on-the-fly to add as many extra nodes as required.
So after 2 weeks of developing and testing, we finally have this feature available. If your Horde front-end supports the “remix” feature. You can send up to 1-6 images to this workflow along with a prompt, and it will try its best to “squash” them all together into one composition. Note that the more images you send, and the larger the prompt, the harder it will be for the model to “retain” all of them in the composition. But it will try its best.
As an example, here’s how the model remixes my own avatar. You’ll notice that the result can understand the general concepts of the image, but can’t follow it exactly as it’s not doing img2img. The blur is probably caused by the need to upscale my original image, which is something I’d like to fix on the next pass.
Original AvatarRemix AvatarLikewise, this is the Haidra logo
Original LogoRemix LogoAnd finally, here’s a remix of both logo and avatar together
Pretty neat, huh?
This ability to send extra source images also lays the groundwork for the Horde to support things like InstantID, which I hope I’ll be able to work on supporting soon enough.
-
The initial deployment of the Stable Cascade (SC) on the AI Horde supported just text2image workflows, but that was just a subset of what this model can do. We still needed to onboard the rest of its capabilities.
One such capability was the “image variations” option, which allows you to send an image to the model, and get a variation of that image, perhaps with extra stuff added in, using the unClip technology. This required quite a bit of work on hordelib so that it uses a completely different ComfyUI workflow but ultimately this was not so much harder than just adding the img2img capabilities to SC.
The larger difficulty came when I wanted to add the feature to remix multiple images together. The problem being that until now the AI Horde only supported sending a single source image and a single source mask, so a varying amount of images was not possible at all.
So to support this, I needed to touch all areas of the AI Horde. The AI Horde had to accept and upload each of them on my R2 bucket and provide individual download links. The SDK had to know to expect and provide methods to download those images in parallel to avoid delays, to the reGen worker had to be able to receive those images and send them to hordelib which should know how to dynamically adjust a comfyUI pipeline on-the-fly to add as many extra nodes as required.
So after 2 weeks of developing and testing, we finally have this feature available. If your Horde front-end supports the “remix” feature. You can send up to 1-6 images to this workflow along with a prompt, and it will try its best to “squash” them all together into one composition. Note that the more images you send, and the larger the prompt, the harder it will be for the model to “retain” all of them in the composition. But it will try its best.
As an example, here’s how the model remixes my own avatar. You’ll notice that the result can understand the general concepts of the image, but can’t follow it exactly as it’s not doing img2img. The blur is probably caused by the need to upscale my original image, which is something I’d like to fix on the next pass.
Original AvatarRemix AvatarLikewise, this is the Haidra logo
Original LogoRemix LogoAnd finally, here’s a remix of both logo and avatar together
Pretty neat, huh?
This ability to send extra source images also lays the groundwork for the Horde to support things like InstantID, which I hope I’ll be able to work on supporting soon enough.
-
The initial deployment of the Stable Cascade (SC) on the AI Horde supported just text2image workflows, but that was just a subset of what this model can do. We still needed to onboard the rest of its capabilities.
One such capability was the “image variations” option, which allows you to send an image to the model, and get a variation of that image, perhaps with extra stuff added in, using the unClip technology. This required quite a bit of work on hordelib so that it uses a completely different ComfyUI workflow but ultimately this was not so much harder than just adding the img2img capabilities to SC.
The larger difficulty came when I wanted to add the feature to remix multiple images together. The problem being that until now the AI Horde only supported sending a single source image and a single source mask, so a varying amount of images was not possible at all.
So to support this, I needed to touch all areas of the AI Horde. The AI Horde had to accept and upload each of them on my R2 bucket and provide individual download links. The SDK had to know to expect and provide methods to download those images in parallel to avoid delays, to the reGen worker had to be able to receive those images and send them to hordelib which should know how to dynamically adjust a comfyUI pipeline on-the-fly to add as many extra nodes as required.
So after 2 weeks of developing and testing, we finally have this feature available. If your Horde front-end supports the “remix” feature. You can send up to 1-6 images to this workflow along with a prompt, and it will try its best to “squash” them all together into one composition. Note that the more images you send, and the larger the prompt, the harder it will be for the model to “retain” all of them in the composition. But it will try its best.
As an example, here’s how the model remixes my own avatar. You’ll notice that the result can understand the general concepts of the image, but can’t follow it exactly as it’s not doing img2img. The blur is probably caused by the need to upscale my original image, which is something I’d like to fix on the next pass.
Original AvatarRemix AvatarLikewise, this is the Haidra logo
Original LogoRemix LogoAnd finally, here’s a remix of both logo and avatar together
Pretty neat, huh?
This ability to send extra source images also lays the groundwork for the Horde to support things like InstantID, which I hope I’ll be able to work on supporting soon enough.
-
A while ago Stability.ai released a new model on a different architecture, that seems to provide very promising results and very fast training: Stable Cascade. I really wished to offer it on the AI Horde so after getting explicit permission from Emad in Reddit PMs (due to its more restrictive license for APIs), I set out to implement it.
Unfortunately the Stable Cascade model and ComfyUI workflow require the use of two different checkpoints, which went against the AI Horde worker paradigm at the time, which expected one file per model, so I had to make multiple changes in a lot of packages which expected this paradigm. The Worker, hordelib, the model reference and its SDK, all of them required tweaking to avoid crashing.
Fortunately, while the changes were complicated, I managed to implement them without much debugging. I did initially run into some troubles with the image quality being garbage, which turned out required ComfyAnon tweaking the implementation on ComfyUI a bit, but once that was done, everything fell in place and now you can use the AI Horde to request Stable Cascade images and therefore check the capability of this model, even if you don’t have 20G VRAM to spare.
You can try it out on Artbot
Alongside Stable Cascade, I thought it’s high time we start expanding our SDXL model selection, so the following models have also been onboarded.
- Juggernaut XL
- Anime Illust Diffusion XL
- Pony Diffusion XL
- Animagine XL
- DreamShaper XL (Lightning version)
We quickly realized that we also need to expand our model reference to better inform people of the requirements for some of these models. For example Pony Diffusion XL doesn’t work unless you set clip_skip to 2, and DreamShaper requires low steps, cfg and specific samplers. If you know to set those settings correctly, you’ll get amazing images, else you get hot garbage. Soon the horde will be warning you when trying to use a model outside its specifications.
Other than that, we haven’t been completely idle. Some other notable achievements in the previous weeks are:
Firstly, the AI Horde now supports an educator role for accounts. If you are an education institution and you want to use one of the AI Horde free tools for the classroom, you can request your account to be set as an educator, which will force all your requests to be SFW and increase your account’s concurrency.
I also spent some time improving the AI Generation of the Mastodon bot @dungeons, so that it gets nicer images for each campaign protagonist. Will admit I had a lot more fun than I should improving the versatility and variability of the generations and tweaking then results for each model. You can see (or follow) the results in the dedicated account replying with those images.
On the worker side, Tazlin has also been very busy improving the efficiency of our generations. We have added now some improvements such as downloading the loras for the next job, while performing the inference for the previous one, or adding more efficiency for those people with more powerful machines.
I’m now hard at work trying to onboard more Stable Cascade capabilities as they are added to ComfyUI and to add support for more advanced workflow capabilities.
-
A while ago Stability.ai released a new model on a different architecture, that seems to provide very promising results and very fast training: Stable Cascade. I really wished to offer it on the AI Horde so after getting explicit permission from Emad in Reddit PMs (due to its more restrictive license for APIs), I set out to implement it.
Unfortunately the Stable Cascade model and ComfyUI workflow require the use of two different checkpoints, which went against the AI Horde worker paradigm at the time, which expected one file per model, so I had to make multiple changes in a lot of packages which expected this paradigm. The Worker, hordelib, the model reference and its SDK, all of them required tweaking to avoid crashing.
Fortunately, while the changes were complicated, I managed to implement them without much debugging. I did initially run into some troubles with the image quality being garbage, which turned out required ComfyAnon tweaking the implementation on ComfyUI a bit, but once that was done, everything fell in place and now you can use the AI Horde to request Stable Cascade images and therefore check the capability of this model, even if you don’t have 20G VRAM to spare.
You can try it out on Artbot
Alongside Stable Cascade, I thought it’s high time we start expanding our SDXL model selection, so the following models have also been onboarded.
- Juggernaut XL
- Anime Illust Diffusion XL
- Pony Diffusion XL
- Animagine XL
- DreamShaper XL (Lightning version)
We quickly realized that we also need to expand our model reference to better inform people of the requirements for some of these models. For example Pony Diffusion XL doesn’t work unless you set clip_skip to 2, and DreamShaper requires low steps, cfg and specific samplers. If you know to set those settings correctly, you’ll get amazing images, else you get hot garbage. Soon the horde will be warning you when trying to use a model outside its specifications.
Other than that, we haven’t been completely idle. Some other notable achievements in the previous weeks are:
Firstly, the AI Horde now supports an educator role for accounts. If you are an education institution and you want to use one of the AI Horde free tools for the classroom, you can request your account to be set as an educator, which will force all your requests to be SFW and increase your account’s concurrency.
I also spent some time improving the AI Generation of the Mastodon bot @dungeons, so that it gets nicer images for each campaign protagonist. Will admit I had a lot more fun than I should improving the versatility and variability of the generations and tweaking then results for each model. You can see (or follow) the results in the dedicated account replying with those images.
On the worker side, Tazlin has also been very busy improving the efficiency of our generations. We have added now some improvements such as downloading the loras for the next job, while performing the inference for the previous one, or adding more efficiency for those people with more powerful machines.
I’m now hard at work trying to onboard more Stable Cascade capabilities as they are added to ComfyUI and to add support for more advanced workflow capabilities.
-
A while ago Stability.ai released a new model on a different architecture, that seems to provide very promising results and very fast training: Stable Cascade. I really wished to offer it on the AI Horde so after getting explicit permission from Emad in Reddit PMs (due to its more restrictive license for APIs), I set out to implement it.
Unfortunately the Stable Cascade model and ComfyUI workflow require the use of two different checkpoints, which went against the AI Horde worker paradigm at the time, which expected one file per model, so I had to make multiple changes in a lot of packages which expected this paradigm. The Worker, hordelib, the model reference and its SDK, all of them required tweaking to avoid crashing.
Fortunately, while the changes were complicated, I managed to implement them without much debugging. I did initially run into some troubles with the image quality being garbage, which turned out required ComfyAnon tweaking the implementation on ComfyUI a bit, but once that was done, everything fell in place and now you can use the AI Horde to request Stable Cascade images and therefore check the capability of this model, even if you don’t have 20G VRAM to spare.
You can try it out on Artbot
Alongside Stable Cascade, I thought it’s high time we start expanding our SDXL model selection, so the following models have also been onboarded.
- Juggernaut XL
- Anime Illust Diffusion XL
- Pony Diffusion XL
- Animagine XL
- DreamShaper XL (Lightning version)
We quickly realized that we also need to expand our model reference to better inform people of the requirements for some of these models. For example Pony Diffusion XL doesn’t work unless you set clip_skip to 2, and DreamShaper requires low steps, cfg and specific samplers. If you know to set those settings correctly, you’ll get amazing images, else you get hot garbage. Soon the horde will be warning you when trying to use a model outside its specifications.
Other than that, we haven’t been completely idle. Some other notable achievements in the previous weeks are:
Firstly, the AI Horde now supports an educator role for accounts. If you are an education institution and you want to use one of the AI Horde free tools for the classroom, you can request your account to be set as an educator, which will force all your requests to be SFW and increase your account’s concurrency.
I also spent some time improving the AI Generation of the Mastodon bot @dungeons, so that it gets nicer images for each campaign protagonist. Will admit I had a lot more fun than I should improving the versatility and variability of the generations and tweaking then results for each model. You can see (or follow) the results in the dedicated account replying with those images.
On the worker side, Tazlin has also been very busy improving the efficiency of our generations. We have added now some improvements such as downloading the loras for the next job, while performing the inference for the previous one, or adding more efficiency for those people with more powerful machines.
I’m now hard at work trying to onboard more Stable Cascade capabilities as they are added to ComfyUI and to add support for more advanced workflow capabilities.
-
A while ago Stability.ai released a new model on a different architecture, that seems to provide very promising results and very fast training: Stable Cascade. I really wished to offer it on the AI Horde so after getting explicit permission from Emad in Reddit PMs (due to its more restrictive license for APIs), I set out to implement it.
Unfortunately the Stable Cascade model and ComfyUI workflow require the use of two different checkpoints, which went against the AI Horde worker paradigm at the time, which expected one file per model, so I had to make multiple changes in a lot of packages which expected this paradigm. The Worker, hordelib, the model reference and its SDK, all of them required tweaking to avoid crashing.
Fortunately, while the changes were complicated, I managed to implement them without much debugging. I did initially run into some troubles with the image quality being garbage, which turned out required ComfyAnon tweaking the implementation on ComfyUI a bit, but once that was done, everything fell in place and now you can use the AI Horde to request Stable Cascade images and therefore check the capability of this model, even if you don’t have 20G VRAM to spare.
You can try it out on Artbot
Alongside Stable Cascade, I thought it’s high time we start expanding our SDXL model selection, so the following models have also been onboarded.
- Juggernaut XL
- Anime Illust Diffusion XL
- Pony Diffusion XL
- Animagine XL
- DreamShaper XL (Lightning version)
We quickly realized that we also need to expand our model reference to better inform people of the requirements for some of these models. For example Pony Diffusion XL doesn’t work unless you set clip_skip to 2, and DreamShaper requires low steps, cfg and specific samplers. If you know to set those settings correctly, you’ll get amazing images, else you get hot garbage. Soon the horde will be warning you when trying to use a model outside its specifications.
Other than that, we haven’t been completely idle. Some other notable achievements in the previous weeks are:
Firstly, the AI Horde now supports an educator role for accounts. If you are an education institution and you want to use one of the AI Horde free tools for the classroom, you can request your account to be set as an educator, which will force all your requests to be SFW and increase your account’s concurrency.
I also spent some time improving the AI Generation of the Mastodon bot @dungeons, so that it gets nicer images for each campaign protagonist. Will admit I had a lot more fun than I should improving the versatility and variability of the generations and tweaking then results for each model. You can see (or follow) the results in the dedicated account replying with those images.
On the worker side, Tazlin has also been very busy improving the efficiency of our generations. We have added now some improvements such as downloading the loras for the next job, while performing the inference for the previous one, or adding more efficiency for those people with more powerful machines.
I’m now hard at work trying to onboard more Stable Cascade capabilities as they are added to ComfyUI and to add support for more advanced workflow capabilities.
-
A while ago Stability.ai released a new model on a different architecture, that seems to provide very promising results and very fast training: Stable Cascade. I really wished to offer it on the AI Horde so after getting explicit permission from Emad in Reddit PMs (due to its more restrictive license for APIs), I set out to implement it.
Unfortunately the Stable Cascade model and ComfyUI workflow require the use of two different checkpoints, which went against the AI Horde worker paradigm at the time, which expected one file per model, so I had to make multiple changes in a lot of packages which expected this paradigm. The Worker, hordelib, the model reference and its SDK, all of them required tweaking to avoid crashing.
Fortunately, while the changes were complicated, I managed to implement them without much debugging. I did initially run into some troubles with the image quality being garbage, which turned out required ComfyAnon tweaking the implementation on ComfyUI a bit, but once that was done, everything fell in place and now you can use the AI Horde to request Stable Cascade images and therefore check the capability of this model, even if you don’t have 20G VRAM to spare.
Alongside Stable Cascade, I thought it’s high time we start expanding our SDXL model selection, so the following models have also been onboarded.
- Juggernaut XL
- Anime Illust Diffusion XL
- Pony Diffusion XL
- Animagine XL
- DreamShaper XL (Lightning version)
We quickly realized that we also need to expand our model reference to better inform people of the requirements for some of these models. For example Pony Diffusion XL doesn’t work unless you set clip_skip to 2, and DreamShaper requires low steps, cfg and specific samplers. If you know to set those settings correctly, you’ll get amazing images, else you get hot garbage. Soon the horde will be warning you when trying to use a model outside its specifications.
Other than that, we haven’t been completely idle. Some other notable achievements in the previous weeks are:
Firstly, the AI Horde now supports an educator role for accounts. If you are an education institution and you want to use one of the AI Horde free tools for the classroom, you can request your account to be set as an educator, which will force all your requests to be SFW and increase your account’s concurrency.
I also spent some time improving the AI Generation of the Mastodon bot @dungeons, so that it gets nicer images for each campaign protagonist. Will admit I had a lot more fun than I should improving the versatility and variability of the generations and tweaking then results for each model. You can see (or follow) the results in the dedicated account replying with those images.
On the worker side, Tazlin has also been very busy improving the efficiency of our generations. We have added now some improvements such as downloading the loras for the next job, while performing the inference for the previous one, or adding more efficiency for those people with more powerful machines.
I’m now hard at work trying to onboard more Stable Cascade capabilities as they are added to ComfyUI and to add support for more advanced workflow capabilities.
-
لجعل محمل الإقلاع (GRUB) في توزيعة Gnu/Linux يتذكر آخر نظام تشغيل قمت بتشغيله تلقائيًا دون الحاجة إلى اختياره يدويًا في كل مرة، يمكنك تعديل إعدادات GRUB باتباع الخطوات التالية:
1. تعديل ملف إعدادات GRUB:
افتح الطرفية (Terminal).
قم بتحرير ملف إعدادات GRUB باستخدام محرر النصوص
nanoأو أي محرر آخر تفضله:sudo nano /etc/default/grubابحث عن السطر التالي:
GRUB_DEFAULT=0قم بتغييره إلى:
GRUB_DEFAULT=savedأضف السطر التالي أسفل السطر السابق:
GRUB_SAVEDEFAULT=trueاحفظ التغييرات وأغلق المحرر (في
nano: اضغطCtrl + X، ثمY، ثمEnter).
2. تحديث GRUB:
- بعد تعديل الملف، قم بتحديث GRUB لتطبيق التغييرات:
sudo update-grub
3. إعادة التشغيل:
- أعد تشغيل النظام، وسيتذكر GRUB الآن آخر نظام تشغيل قمت بتشغيله.
ملاحظات إضافية:
- إذا كنت تستخدم systemd-boot بدلاً من GRUB، يمكنك تعديل إعداداته عبر ملف
/boot/loader/loader.conf. - تأكد من أن التغييرات تطبق بشكل صحيح عن طريق إعادة التشغيل واختبار النظام.
#Manjaro #GRUB #Linux #إعدادات_GRUB #توزيعات_لينكس #نظام_تشغيل #إقلاع_تلقائي #إدارة_الإقلاع #تخصيص_النظام #نصائح_لينكس #Manjaro #GRUB #Linux #BootLoader #LinuxTips #SystemCustomization #AutoBoot #LinuxTweaks #TechTips #OpenSource #Gnutux
-
"Un noto studioso del fascismo, Robert Paxton, ha scritto nel suo libro “The Anatomy of Fascism” (L'anatomia del fascismo) che i movimenti fascisti non possono mai nascere senza l'aiuto della gente comune, anche della brava gente."
✍️ Zvi Bar'el
🧵1/8#CeasefireNOW #SanctionIsraelNOW
#ZionistTerrorism
#GazaGenocide #Gaza #Israel #IsraeliOccupation #PACE #2gennaio#CeasefireNOW #SanctionIsraelNOW
#ZionistTerrorism
#GazaGenocide #Gaza #Israel #IsraeliOccupation #PACE #2gennaio -
CW: NSFW 18+ Nudity
-
CW: NSFW 18+ Nudity
