Really like the ability to run multiple models simultaneously in Open Web AI, which helps me validate how a "lesser" model such as Deepseek-r1:16b compares against OpenAI GPT-4o - and it's pretty good actually!
Really like the ability to run multiple models simultaneously in Open Web AI, which helps me validate how a "lesser" model such as Deepseek-r1:16b compares against OpenAI GPT-4o - and it's pretty good actually!
Try asking the models wrongthink questions to check for censorship and bias, absolute or relative
