@OmAI_lab@OmAI_lab This is fantastic! When will the weights for VLX-Seek-1.5-3B be open-sourced? The 10B version is excellent, but the resource requirements are so high that it’s quite daunting.
@thsottiaux My experience with “dot”: Powered by the Astra engine, it allows for voice-based interaction. During the confirmation phase, it articulates the user’s request very clearly—yet, once it proceeds to execute the task, the actual outcome is a complete mess.
@thsottiaux What’s going on today? I keep getting the message: "Selected model is at capacity. Please try a different model." Is this related to an upcoming reset, or are they getting ready to release a new model? Does anyone in the know have any inside scoop?
@thsottiaux@JustLingonberry@sama I don’t see it that way; one only needs to look at the growth in OpenAI’s user base over the past three months—all of which can be credited to Tibo. In my view, Tibo is a sincere and outstanding individual.
@thsottiaux I really have to vent about the speed issues with GPT-6; even in "fast" mode, it moves at a snail's pace—it takes over an hour to complete even a simple task.
How about you guys?
@thsottiaux I feel like my quota gets used up really fast after resetting the weekly limit with a top-up card—has anyone else experienced this? (Pro x5 user)
@thsottiaux That’s somewhat distorting the facts; sub2api users wouldn't even pay attention to what you're saying—they don't care about resets. Genuine user experiences don't lie.