AI & RoboticsNews

Stability AI sows gen AI discord with Stable Artisan

Stability AI is finally bringing its generative AI technology for text and video content to Discord.

The new Stable Artisan Discord bot service officially launched today, enabling users to generate images with Stable Diffusion 3 (SD3) and video with Stable Video Diffusion (SVD) via a Discord bot. The idea of using Discord as an interface is one that rival gen AI image service Midjourney has had since it first debuted in 2022.

Stability AI has been providing access to its core Stable Diffusion models via an API that developers could access. For the core model, the weights were also available on HuggingFace. Until February 22 of this year, Stable Diffusion was also available via the Clipdrop web service, which is now owned by gen AI startup Jasper. Stable Video on the other hand is still available via the StableVideo.com website which is owned and operated by Stability AI.

“The initial launch of Stable Artisan will provide us with the chance to monitor user behavior and to verify the current functionality and operation of the Discord bot are up to our standards before we start adding more features,” Christian Laforte, co-CEO and CTO of Stability AI told VentureBeat.

Stable Artisan is a paid service with a 3-day free trial to start.

The service has a range of plans from the standard plan for $9 which provides 900 credits per month, up to the $99 premium plan that provides 1,200 credits a month. 

Stability AI’s API uses a credit system to determine the value of an image generation. Each SD3 image is 6.5 credits, while a generation with the older but still capable Stable Diffusion image core model is only 3 credits. Video is priced at 20 credits per generation.

Laforte said that to access Stable Artisan, users can create an account on the Stability AI  website and then join the company’s Discord server to get started. 

Stability AI also has a membership system that it announced at the end of 2023. Laforte said that the Stable Artisan service is not currently linked to the Stability AI Membership.

Stable Artisan isn’t just about basic image generation. It also has several editing and customization capabilities as well. The customization capabilities are somewhat similar to what Stability AI had previously offered when it operated the Clipdrop web service.

Among the features are:

While the new Stable Artisan service provides a more accessible way for many users to access Stability AI image generation models, there is still more to come.

For one, at least as of today, the Stable Artisan does not provide access to Stable Audio. Last month Stability AI debuted Stable Audio 2.0, which is currently available via API as well as the StableAudio.com website. Stability AI also has several other models for code and text generation including Stable Code and Stable LM, neither of which are available on Stable Artisan either.

Stability AI is however working on a larger conversational chatbot idea known as Stable Assistant, which was announced in beta when SD3 became available last month. Stable Assistant is a web-based friendly chatbot that uses Stability AI’s text-to-image and large language model (LLM) technologies to assist users with a range of tasks over natural language conversations.

“They are both applications that are actively developed by Stability AI and address different groups of users,” Laforte said.

Join us in returning to NYC on June 5th to collaborate with executive leaders in exploring comprehensive methods for auditing AI models regarding bias, performance, and ethical compliance across diverse organizations. Find out how you can attend here.


Stability AI is finally bringing its generative AI technology for text and video content to Discord.

The new Stable Artisan Discord bot service officially launched today, enabling users to generate images with Stable Diffusion 3 (SD3) and video with Stable Video Diffusion (SVD) via a Discord bot. The idea of using Discord as an interface is one that rival gen AI image service Midjourney has had since it first debuted in 2022.

Stability AI has been providing access to its core Stable Diffusion models via an API that developers could access. For the core model, the weights were also available on HuggingFace. Until February 22 of this year, Stable Diffusion was also available via the Clipdrop web service, which is now owned by gen AI startup Jasper. Stable Video on the other hand is still available via the StableVideo.com website which is owned and operated by Stability AI.

“The initial launch of Stable Artisan will provide us with the chance to monitor user behavior and to verify the current functionality and operation of the Discord bot are up to our standards before we start adding more features,” Christian Laforte, co-CEO and CTO of Stability AI told VentureBeat.

VB Event

The AI Impact Tour: The AI Audit

Join us as we return to NYC on June 5th to engage with top executive leaders, delving into strategies for auditing AI models to ensure fairness, optimal performance, and ethical compliance across diverse organizations. Secure your attendance for this exclusive invite-only event.


Request an invite

You don’t need to be a member to use, but it will cost $

Stable Artisan is a paid service with a 3-day free trial to start.

The service has a range of plans from the standard plan for $9 which provides 900 credits per month, up to the $99 premium plan that provides 1,200 credits a month. 

Stability AI’s API uses a credit system to determine the value of an image generation. Each SD3 image is 6.5 credits, while a generation with the older but still capable Stable Diffusion image core model is only 3 credits. Video is priced at 20 credits per generation.

Laforte said that to access Stable Artisan, users can create an account on the Stability AI  website and then join the company’s Discord server to get started. 

Stability AI also has a membership system that it announced at the end of 2023. Laforte said that the Stable Artisan service is not currently linked to the Stability AI Membership.

What can you do with the Stable Artisan Discord bot?

Stable Artisan isn’t just about basic image generation. It also has several editing and customization capabilities as well. The customization capabilities are somewhat similar to what Stability AI had previously offered when it operated the Clipdrop web service.

Among the features are:

  • Search and Replace. This inpainting service allows users to identify an object within an image using simple language prompts. The identified object is automatically segmented and replaced with the requested object specified in the prompt, without requiring a mask.
  • Remove Background. This feature removes the background while preserving the foreground content.
  • Creative Upscale: Users can take low-resolution or low-quality images and upscale them to up to 4K resolution.
  • Outpaint: The Outpaint service extends the content of an image by inserting additional elements in any direction, filling the expanded space seamlessly.
  • Control Sketch: This enables users to transform rough hand-drawn sketches into high-quality images with precise control. 
  • Control Structure: This feature allows users to generate images while maintaining the structure of an input image.

Stable Assistant is up next

While the new Stable Artisan service provides a more accessible way for many users to access Stability AI image generation models, there is still more to come.

For one, at least as of today, the Stable Artisan does not provide access to Stable Audio. Last month Stability AI debuted Stable Audio 2.0, which is currently available via API as well as the StableAudio.com website. Stability AI also has several other models for code and text generation including Stable Code and Stable LM, neither of which are available on Stable Artisan either.

Stability AI is however working on a larger conversational chatbot idea known as Stable Assistant, which was announced in beta when SD3 became available last month. Stable Assistant is a web-based friendly chatbot that uses Stability AI’s text-to-image and large language model (LLM) technologies to assist users with a range of tasks over natural language conversations.

“They are both applications that are actively developed by Stability AI and address different groups of users,” Laforte said.


Author: Sean Michael Kerner
Source: Venturebeat
Reviewed By: Editorial Team

Related posts
AI & RoboticsNews

DeepSeek’s first reasoning model R1-Lite-Preview turns heads, beating OpenAI o1 performance

AI & RoboticsNews

Snowflake beats Databricks to integrating Claude 3.5 directly

AI & RoboticsNews

OpenScholar: The open-source A.I. that’s outperforming GPT-4o in scientific research

DefenseNews

US Army fires Precision Strike Missile in salvo shot for first time

Sign up for our Newsletter and
stay informed!