With Valve recently confirming that they will host games that contain assets generated by AI, I assume there are now a variety of tools and vendors looking to enter the mainstream. And adding to this thread is god's work.
When I looked into this a year ago, sites like midjourney were positioned as toys and they were explicitly claiming ownership over any generated images while refusing to grant a commercial license.
What I am looking for is a tool that will generate images I have rights to use that doesn't require me to provide the training data. Has anyone been paying attention to this space? Any opinions about what's available now? Anyone using Scenario for a project? Better to just run a tool on your own video card?
With Valve recently confirming that they will host games that contain assets generated by AI, I assume there are now a variety of tools and vendors looking to enter the mainstream. And adding to this thread is god's work.
When I looked into this a year ago, sites like midjourney were positioned as toys and they were explicitly claiming ownership over any generated images while refusing to grant a commercial license.
What I am looking for is a tool that will generate images I have rights to use that doesn't require me to provide the training data. Has anyone been paying attention to this space? Any opinions about what's available now? Anyone using Scenario for a project? Better to just run a tool on your own video card?
All SaaS services suck shit. Running Stable Diffusion locally trumps all of it.
If you don't wanna futz with training data but you need something specific look at Control Net / IP Adapter. Segments & Open Pose & After Detailer are also must have toolbox items. Everything else is fluff or outdated.
With Valve recently confirming that they will host games that contain assets generated by AI, I assume there are now a variety of tools and vendors looking to enter the mainstream. And adding to this thread is god's work.
When I looked into this a year ago, sites like midjourney were positioned as toys and they were explicitly claiming ownership over any generated images while refusing to grant a commercial license.
What I am looking for is a tool that will generate images I have rights to use that doesn't require me to provide the training data. Has anyone been paying attention to this space? Any opinions about what's available now? Anyone using Scenario for a project? Better to just run a tool on your own video card?
All SaaS services suck shit. Running Stable Diffusion locally trumps all of it.
If you don't wanna futz with training data but you need something specific look at Control Net / IP Adapter. Segments & Open Pose & After Detailer are also must have toolbox items. Everything else is fluff or outdated.
I agree that applications trump apps and I have the hardware to do this.
Can you point me at a quick start guide and suggest a model that is not encumbered by licensing/ownership issues?
After you paste the plugin into the folder, it's pretty much a 1-click install. Just make sure you have the space and a non-potato video card - preferably Nvidia.