@bridgemindai: People are quick to say Grok 4.5 sucks as a coding model, but they are ignoring the math. At just 1.5 trillion parameters, the model is actually incredibly small compared to the competition. They have the most compute out of anyone, but they simply have not scaled up their models yet. You cannot judge a 1.5 trillion parameter model against GPT 5.6 or Opus 5. Those models are running on multiple, multiple trillion parameters. What do you think will happen when Grok finally scales up their parameters? #vibecoding #agenticcoding #aicoding #grok #llm #machinelearning #buildthefuture