Graydient is a suite of AI tools for creatives and enthusiasts. We give you easy and convenient online access to large language models and fine-tuned image models that can do things that popular models like Midjourney and Dall-e cannot. Our AI models are trained by the community, and you can use our software to train models too without any technical knowledge. We provide free guides and videos, and there is optional 1-on-1 coaching also available.

You can try five free chat and image creation bots on our Examples page. To unlock the full creative suite and more LLM options, please upgrade to the full version.

The Starter and Plus plan come with the most popular version of Llama LLM, and thousands of fine-tune image models. To unlock every LLM model and beta models, upgrade to Plus. The difference between the three plans are the GPU speeds, though in general we try to deliver a batch of images under a minute.

There are no daily limits, most people create hundreds of images a day.

The max number of images that you can create a time depends on the resolution. Modern models like Flux can be as big as 22GB and that’s not counting the clip models, vae, and all of our platform software. We try to squeeze as many in one shot as possible for you, given those variables.

Our goal is to return your prompt in under a minute, so we balance this depending on the resolution only.  You can stack 3 loras (this increases the vram load) and inversions and use SD1.5 or SD Pony or SDXL or Stable Diffusion 3, and the same rules apply:

A real world example: You can get up to 9 images per prompt at 512×768, or 5 images per prompt at 1024×1024, or 3 images per prompt at 1300×1300.  You can then upscale in a second step beyond 4k.  

After you queue the images and get a confirmation, you can immediately queue more images before you receive them.

Yes, we guarantee high-speed image creation and compatibility on all devices. All you need is a web browser. We generate and store your images all on our compute cloud, meaning we’re not draining your battery or filling up your hard drive. Even if your device is lost or breaks, all of your prompts and images are still safe.

Our mission is to provide you with a service that is faster than a top-of-the-line $10,000 computer, leased to you at a tiny fraction of the cost.

Our hardware is extraordinary: We use the latest AMD and Intel Chips for full PCI lane efficiency, 250GB+ of RAM, large solid state NVME drives.

Unlike our competitors, we aren’t leasing random people’s basement computers while they’re playing Counterstrike or mining Bitcoins. All of our hardware is strictly dedicated to A.I.

Graydient owns or leases A100s, A6000s, RTX 4090s, and upgrades regularly. We do not partition the GPUs, give you the full might of each server. We also support LCM which can deliver 1024×1024 SDXL images under 8 seconds.

Our service is fast, we guarantee it.

Yes, that’s actually one of the most popular activities on our software right now, and there are many groups sharing their tips and prompts. We support three types: Training an SDXL Lora with your own images, consistent character creation through IP Adapters and Loras in our Stable2Go or PirateDiffusion creation systems, and it’s possible to connect them to an LLM to chat with them and send selfies using our PollyGPT software.

You can even do it from your smartphone. We provide detailed tutorials on how to create an A.I. persona, change their clothes or backgrounds, and maintain a persistent look.

Yes, your content is protected with 2FA password logins and is not published or shared in a feed unless you create a public “space” or share on groups. For the safety of the community, there are random anonymous checks for illegal activities.

Yes, this is called creating a “model”, in our case a mini-model called a LoRa (low rank adaptation). You just need at least five clear photos of your face in different lighting and we’ll help you with the rest. Unlike other services, we don’t send you a dozen pictures and run with your cash. Plus members can retrain the model an unlimited number of times as your look changes or if it’s not quite right, and create as many images and situations as you’d like. There’s no toll booth.

None at all. We’re a small family company, and our moms raised us right. You won’t find any fine print. Our service is truly unlimited, and if you fill your drive you can simply delete images or change plans. We don’t nickel and dime our customers. You can easily download your images, too.

We stand by our products. If you’re not satisfied for whatever reason, we’ll promptly refund you asap. All we ask is an opportunity to help you with any issue you’re experiencing first: We are online 24/7 and available by chat – click the VIP link on your dashboard to chat directly with our dev team.

It’s not applicable, because Graydient is the most convenient way to access Open Source A.I. models, optimized for API developers, your web browser and Telegram. We are the software part of the ecosystem, not the models part.

Graydient does not create models or scrape the web to train images. Additionally, any photos that enter the Graydient platform are not sold or trained for other uses. We take our customer’s privacy very seriously.

Persons who do not want to appear in A.I. models should opt out with Stability AI, the creators of the Stable Diffusion model (or) similar creators.

Graydient co-founders Yanier and Thomas built websites like Destructoid, Upcomer, and Classic dot com. We’ve worked as professional web developers for over 20 years. This is not our first rodeo, and take great pride and care of our community. We built Graydient with our own money, because we couldn’t quite find anything like it. We hope to grow this into something wonderful for the Stable Diffusion community at large. We also employ really lovely moderators, podcasters, and sponsor Stable Diffusion model makers occasionally.